Upload 28 files
Browse files- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-600/adapter_config.json +16 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-600/adapter_model.bin +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-600/optimizer.pt +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-600/rng_state.pth +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-600/scheduler.pt +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-600/trainer_state.json +220 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-600/training_args.bin +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-8800/adapter_config.json +16 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-8800/adapter_model.bin +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-8800/optimizer.pt +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-8800/rng_state.pth +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-8800/scheduler.pt +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-8800/trainer_state.json +3008 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-8800/training_args.bin +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-9000/adapter_config.json +16 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-9000/adapter_model.bin +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-9000/optimizer.pt +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-9000/rng_state.pth +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-9000/scheduler.pt +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-9000/trainer_state.json +3076 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-9000/training_args.bin +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail_6000_samples/adapter_config.json +16 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail_6000_samples/adapter_model.bin +3 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail_6000_samples/special_tokens_map.json +6 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail_6000_samples/tokenizer.json +0 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail_6000_samples/tokenizer_config.json +10 -0
- redpj7B-lora-cnn-dailymail_6000_samples/results/stdout.txt +0 -0
- redpj7B-lora-cnn-dailymail_6000_samples/script_fine_tuning.py +170 -0
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-600/adapter_config.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"base_model_name_or_path": "/domino/edv/afs-mrmc-data-store-rw/innovation/hf/RedPajama-INCITE-7B-Base",
|
3 |
+
"bias": "none",
|
4 |
+
"fan_in_fan_out": false,
|
5 |
+
"inference_mode": true,
|
6 |
+
"init_lora_weights": true,
|
7 |
+
"lora_alpha": 16,
|
8 |
+
"lora_dropout": 0.05,
|
9 |
+
"modules_to_save": null,
|
10 |
+
"peft_type": "LORA",
|
11 |
+
"r": 8,
|
12 |
+
"target_modules": [
|
13 |
+
"query_key_value"
|
14 |
+
],
|
15 |
+
"task_type": "CAUSAL_LM"
|
16 |
+
}
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-600/adapter_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7004b69a5e5338ee5dbc682271079d0cf6750a64f3dd06f0dabf0a4c8129f41b
|
3 |
+
size 16800753
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-600/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e30ce31dcb0eb6fc197fa6178337af09855e350ceabc29748e3bfad2865b0828
|
3 |
+
size 33592261
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-600/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bf3c8c6a24cf587ccbaf1b3ebf006c31756fbbb284486b45c4447d10a3da99c3
|
3 |
+
size 14575
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-600/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ff6d8dcf74529ca71baf62714afafbea88dc48c81b1277347aebdfdd87c4533d
|
3 |
+
size 627
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-600/trainer_state.json
ADDED
@@ -0,0 +1,220 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 1.9635850191116333,
|
3 |
+
"best_model_checkpoint": "./results/redpj7B-lora-cnn-dailymail-results_fine_tune_test/checkpoint-600",
|
4 |
+
"epoch": 0.19900497512437812,
|
5 |
+
"global_step": 600,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.01,
|
12 |
+
"learning_rate": 0.00027082228116710874,
|
13 |
+
"loss": 1.8654,
|
14 |
+
"step": 20
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.01,
|
18 |
+
"learning_rate": 0.0002681697612732095,
|
19 |
+
"loss": 1.8124,
|
20 |
+
"step": 40
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.02,
|
24 |
+
"learning_rate": 0.0002655172413793103,
|
25 |
+
"loss": 1.8231,
|
26 |
+
"step": 60
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.03,
|
30 |
+
"learning_rate": 0.0002628647214854111,
|
31 |
+
"loss": 1.7818,
|
32 |
+
"step": 80
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.03,
|
36 |
+
"learning_rate": 0.00026021220159151194,
|
37 |
+
"loss": 1.8446,
|
38 |
+
"step": 100
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.04,
|
42 |
+
"learning_rate": 0.0002575596816976127,
|
43 |
+
"loss": 1.9001,
|
44 |
+
"step": 120
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.05,
|
48 |
+
"learning_rate": 0.0002549071618037135,
|
49 |
+
"loss": 1.8073,
|
50 |
+
"step": 140
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.05,
|
54 |
+
"learning_rate": 0.0002522546419098143,
|
55 |
+
"loss": 1.8506,
|
56 |
+
"step": 160
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.06,
|
60 |
+
"learning_rate": 0.0002496021220159151,
|
61 |
+
"loss": 1.8471,
|
62 |
+
"step": 180
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.07,
|
66 |
+
"learning_rate": 0.0002469496021220159,
|
67 |
+
"loss": 1.8647,
|
68 |
+
"step": 200
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.07,
|
72 |
+
"eval_loss": 1.966022253036499,
|
73 |
+
"eval_runtime": 12002.3952,
|
74 |
+
"eval_samples_per_second": 1.114,
|
75 |
+
"eval_steps_per_second": 0.139,
|
76 |
+
"step": 200
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 0.07,
|
80 |
+
"learning_rate": 0.00024429708222811666,
|
81 |
+
"loss": 1.8578,
|
82 |
+
"step": 220
|
83 |
+
},
|
84 |
+
{
|
85 |
+
"epoch": 0.08,
|
86 |
+
"learning_rate": 0.0002416445623342175,
|
87 |
+
"loss": 1.8329,
|
88 |
+
"step": 240
|
89 |
+
},
|
90 |
+
{
|
91 |
+
"epoch": 0.09,
|
92 |
+
"learning_rate": 0.0002389920424403183,
|
93 |
+
"loss": 1.8119,
|
94 |
+
"step": 260
|
95 |
+
},
|
96 |
+
{
|
97 |
+
"epoch": 0.09,
|
98 |
+
"learning_rate": 0.0002363395225464191,
|
99 |
+
"loss": 1.8884,
|
100 |
+
"step": 280
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.1,
|
104 |
+
"learning_rate": 0.00023368700265251986,
|
105 |
+
"loss": 1.9077,
|
106 |
+
"step": 300
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 0.11,
|
110 |
+
"learning_rate": 0.00023103448275862065,
|
111 |
+
"loss": 1.8092,
|
112 |
+
"step": 320
|
113 |
+
},
|
114 |
+
{
|
115 |
+
"epoch": 0.11,
|
116 |
+
"learning_rate": 0.00022838196286472146,
|
117 |
+
"loss": 1.9237,
|
118 |
+
"step": 340
|
119 |
+
},
|
120 |
+
{
|
121 |
+
"epoch": 0.12,
|
122 |
+
"learning_rate": 0.00022572944297082225,
|
123 |
+
"loss": 1.8701,
|
124 |
+
"step": 360
|
125 |
+
},
|
126 |
+
{
|
127 |
+
"epoch": 0.13,
|
128 |
+
"learning_rate": 0.00022307692307692306,
|
129 |
+
"loss": 1.8933,
|
130 |
+
"step": 380
|
131 |
+
},
|
132 |
+
{
|
133 |
+
"epoch": 0.13,
|
134 |
+
"learning_rate": 0.00022042440318302385,
|
135 |
+
"loss": 1.8453,
|
136 |
+
"step": 400
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.13,
|
140 |
+
"eval_loss": 1.9667036533355713,
|
141 |
+
"eval_runtime": 11991.8036,
|
142 |
+
"eval_samples_per_second": 1.115,
|
143 |
+
"eval_steps_per_second": 0.139,
|
144 |
+
"step": 400
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 0.14,
|
148 |
+
"learning_rate": 0.00021777188328912466,
|
149 |
+
"loss": 1.8272,
|
150 |
+
"step": 420
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 0.15,
|
154 |
+
"learning_rate": 0.00021511936339522545,
|
155 |
+
"loss": 1.8149,
|
156 |
+
"step": 440
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.15,
|
160 |
+
"learning_rate": 0.00021246684350132626,
|
161 |
+
"loss": 1.8565,
|
162 |
+
"step": 460
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 0.16,
|
166 |
+
"learning_rate": 0.00020981432360742705,
|
167 |
+
"loss": 1.8139,
|
168 |
+
"step": 480
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 0.17,
|
172 |
+
"learning_rate": 0.0002071618037135278,
|
173 |
+
"loss": 1.802,
|
174 |
+
"step": 500
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 0.17,
|
178 |
+
"learning_rate": 0.00020450928381962862,
|
179 |
+
"loss": 1.8795,
|
180 |
+
"step": 520
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 0.18,
|
184 |
+
"learning_rate": 0.0002018567639257294,
|
185 |
+
"loss": 1.8621,
|
186 |
+
"step": 540
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 0.19,
|
190 |
+
"learning_rate": 0.00019920424403183022,
|
191 |
+
"loss": 1.8363,
|
192 |
+
"step": 560
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 0.19,
|
196 |
+
"learning_rate": 0.000196551724137931,
|
197 |
+
"loss": 1.8746,
|
198 |
+
"step": 580
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.2,
|
202 |
+
"learning_rate": 0.00019389920424403182,
|
203 |
+
"loss": 1.7763,
|
204 |
+
"step": 600
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 0.2,
|
208 |
+
"eval_loss": 1.9635850191116333,
|
209 |
+
"eval_runtime": 11976.5234,
|
210 |
+
"eval_samples_per_second": 1.116,
|
211 |
+
"eval_steps_per_second": 0.14,
|
212 |
+
"step": 600
|
213 |
+
}
|
214 |
+
],
|
215 |
+
"max_steps": 9045,
|
216 |
+
"num_train_epochs": 3,
|
217 |
+
"total_flos": 3.74219846516736e+16,
|
218 |
+
"trial_name": null,
|
219 |
+
"trial_params": null
|
220 |
+
}
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-600/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f5a78612ea8930d68eba4cb53d62254ccf547582e754aa049d169c3c11dd5fe4
|
3 |
+
size 4027
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-8800/adapter_config.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"base_model_name_or_path": "/domino/edv/afs-mrmc-data-store-rw/innovation/hf/RedPajama-INCITE-7B-Base",
|
3 |
+
"bias": "none",
|
4 |
+
"fan_in_fan_out": false,
|
5 |
+
"inference_mode": true,
|
6 |
+
"init_lora_weights": true,
|
7 |
+
"lora_alpha": 16,
|
8 |
+
"lora_dropout": 0.05,
|
9 |
+
"modules_to_save": null,
|
10 |
+
"peft_type": "LORA",
|
11 |
+
"r": 8,
|
12 |
+
"target_modules": [
|
13 |
+
"query_key_value"
|
14 |
+
],
|
15 |
+
"task_type": "CAUSAL_LM"
|
16 |
+
}
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-8800/adapter_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:36c95157caaa0e8a49516175a38f9e3dfad6634df6c2d7fc47b2e3298cf4d68e
|
3 |
+
size 16800753
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-8800/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f45578cb0861cd43c16230fed120f8164d6858bd1387fe11181131bf8618591c
|
3 |
+
size 33592261
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-8800/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c4113cbc0a2963d4c364a9d22be5bb0998af83f60ca6808727b9e33650e173f6
|
3 |
+
size 14575
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-8800/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c67c61a9d3974b17ff5914527d23045fa60ef4e1f1e7017cc41f7ea5686f8e9a
|
3 |
+
size 627
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-8800/trainer_state.json
ADDED
@@ -0,0 +1,3008 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 1.9635850191116333,
|
3 |
+
"best_model_checkpoint": "./results/redpj7B-lora-cnn-dailymail-results_fine_tune_test/checkpoint-600",
|
4 |
+
"epoch": 2.9187396351575456,
|
5 |
+
"global_step": 8800,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.01,
|
12 |
+
"learning_rate": 0.00027082228116710874,
|
13 |
+
"loss": 1.8654,
|
14 |
+
"step": 20
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.01,
|
18 |
+
"learning_rate": 0.0002681697612732095,
|
19 |
+
"loss": 1.8124,
|
20 |
+
"step": 40
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.02,
|
24 |
+
"learning_rate": 0.0002655172413793103,
|
25 |
+
"loss": 1.8231,
|
26 |
+
"step": 60
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.03,
|
30 |
+
"learning_rate": 0.0002628647214854111,
|
31 |
+
"loss": 1.7818,
|
32 |
+
"step": 80
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.03,
|
36 |
+
"learning_rate": 0.00026021220159151194,
|
37 |
+
"loss": 1.8446,
|
38 |
+
"step": 100
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.04,
|
42 |
+
"learning_rate": 0.0002575596816976127,
|
43 |
+
"loss": 1.9001,
|
44 |
+
"step": 120
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.05,
|
48 |
+
"learning_rate": 0.0002549071618037135,
|
49 |
+
"loss": 1.8073,
|
50 |
+
"step": 140
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.05,
|
54 |
+
"learning_rate": 0.0002522546419098143,
|
55 |
+
"loss": 1.8506,
|
56 |
+
"step": 160
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.06,
|
60 |
+
"learning_rate": 0.0002496021220159151,
|
61 |
+
"loss": 1.8471,
|
62 |
+
"step": 180
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.07,
|
66 |
+
"learning_rate": 0.0002469496021220159,
|
67 |
+
"loss": 1.8647,
|
68 |
+
"step": 200
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.07,
|
72 |
+
"eval_loss": 1.966022253036499,
|
73 |
+
"eval_runtime": 12002.3952,
|
74 |
+
"eval_samples_per_second": 1.114,
|
75 |
+
"eval_steps_per_second": 0.139,
|
76 |
+
"step": 200
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 0.07,
|
80 |
+
"learning_rate": 0.00024429708222811666,
|
81 |
+
"loss": 1.8578,
|
82 |
+
"step": 220
|
83 |
+
},
|
84 |
+
{
|
85 |
+
"epoch": 0.08,
|
86 |
+
"learning_rate": 0.0002416445623342175,
|
87 |
+
"loss": 1.8329,
|
88 |
+
"step": 240
|
89 |
+
},
|
90 |
+
{
|
91 |
+
"epoch": 0.09,
|
92 |
+
"learning_rate": 0.0002389920424403183,
|
93 |
+
"loss": 1.8119,
|
94 |
+
"step": 260
|
95 |
+
},
|
96 |
+
{
|
97 |
+
"epoch": 0.09,
|
98 |
+
"learning_rate": 0.0002363395225464191,
|
99 |
+
"loss": 1.8884,
|
100 |
+
"step": 280
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.1,
|
104 |
+
"learning_rate": 0.00023368700265251986,
|
105 |
+
"loss": 1.9077,
|
106 |
+
"step": 300
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 0.11,
|
110 |
+
"learning_rate": 0.00023103448275862065,
|
111 |
+
"loss": 1.8092,
|
112 |
+
"step": 320
|
113 |
+
},
|
114 |
+
{
|
115 |
+
"epoch": 0.11,
|
116 |
+
"learning_rate": 0.00022838196286472146,
|
117 |
+
"loss": 1.9237,
|
118 |
+
"step": 340
|
119 |
+
},
|
120 |
+
{
|
121 |
+
"epoch": 0.12,
|
122 |
+
"learning_rate": 0.00022572944297082225,
|
123 |
+
"loss": 1.8701,
|
124 |
+
"step": 360
|
125 |
+
},
|
126 |
+
{
|
127 |
+
"epoch": 0.13,
|
128 |
+
"learning_rate": 0.00022307692307692306,
|
129 |
+
"loss": 1.8933,
|
130 |
+
"step": 380
|
131 |
+
},
|
132 |
+
{
|
133 |
+
"epoch": 0.13,
|
134 |
+
"learning_rate": 0.00022042440318302385,
|
135 |
+
"loss": 1.8453,
|
136 |
+
"step": 400
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.13,
|
140 |
+
"eval_loss": 1.9667036533355713,
|
141 |
+
"eval_runtime": 11991.8036,
|
142 |
+
"eval_samples_per_second": 1.115,
|
143 |
+
"eval_steps_per_second": 0.139,
|
144 |
+
"step": 400
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 0.14,
|
148 |
+
"learning_rate": 0.00021777188328912466,
|
149 |
+
"loss": 1.8272,
|
150 |
+
"step": 420
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 0.15,
|
154 |
+
"learning_rate": 0.00021511936339522545,
|
155 |
+
"loss": 1.8149,
|
156 |
+
"step": 440
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.15,
|
160 |
+
"learning_rate": 0.00021246684350132626,
|
161 |
+
"loss": 1.8565,
|
162 |
+
"step": 460
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 0.16,
|
166 |
+
"learning_rate": 0.00020981432360742705,
|
167 |
+
"loss": 1.8139,
|
168 |
+
"step": 480
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 0.17,
|
172 |
+
"learning_rate": 0.0002071618037135278,
|
173 |
+
"loss": 1.802,
|
174 |
+
"step": 500
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 0.17,
|
178 |
+
"learning_rate": 0.00020450928381962862,
|
179 |
+
"loss": 1.8795,
|
180 |
+
"step": 520
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 0.18,
|
184 |
+
"learning_rate": 0.0002018567639257294,
|
185 |
+
"loss": 1.8621,
|
186 |
+
"step": 540
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 0.19,
|
190 |
+
"learning_rate": 0.00019920424403183022,
|
191 |
+
"loss": 1.8363,
|
192 |
+
"step": 560
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 0.19,
|
196 |
+
"learning_rate": 0.000196551724137931,
|
197 |
+
"loss": 1.8746,
|
198 |
+
"step": 580
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.2,
|
202 |
+
"learning_rate": 0.00019389920424403182,
|
203 |
+
"loss": 1.7763,
|
204 |
+
"step": 600
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 0.2,
|
208 |
+
"eval_loss": 1.9635850191116333,
|
209 |
+
"eval_runtime": 11976.5234,
|
210 |
+
"eval_samples_per_second": 1.116,
|
211 |
+
"eval_steps_per_second": 0.14,
|
212 |
+
"step": 600
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.21,
|
216 |
+
"learning_rate": 0.0001912466843501326,
|
217 |
+
"loss": 1.8423,
|
218 |
+
"step": 620
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 0.21,
|
222 |
+
"learning_rate": 0.00018859416445623343,
|
223 |
+
"loss": 1.8998,
|
224 |
+
"step": 640
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 0.22,
|
228 |
+
"learning_rate": 0.0001859416445623342,
|
229 |
+
"loss": 1.8131,
|
230 |
+
"step": 660
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 0.23,
|
234 |
+
"learning_rate": 0.00018328912466843497,
|
235 |
+
"loss": 1.8656,
|
236 |
+
"step": 680
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.23,
|
240 |
+
"learning_rate": 0.00018063660477453579,
|
241 |
+
"loss": 1.8181,
|
242 |
+
"step": 700
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 0.24,
|
246 |
+
"learning_rate": 0.00017798408488063657,
|
247 |
+
"loss": 1.8425,
|
248 |
+
"step": 720
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.25,
|
252 |
+
"learning_rate": 0.00017533156498673739,
|
253 |
+
"loss": 1.7883,
|
254 |
+
"step": 740
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.25,
|
258 |
+
"learning_rate": 0.00017267904509283817,
|
259 |
+
"loss": 1.8469,
|
260 |
+
"step": 760
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 0.26,
|
264 |
+
"learning_rate": 0.000170026525198939,
|
265 |
+
"loss": 1.8447,
|
266 |
+
"step": 780
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 0.27,
|
270 |
+
"learning_rate": 0.00016737400530503977,
|
271 |
+
"loss": 1.7872,
|
272 |
+
"step": 800
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 0.27,
|
276 |
+
"eval_loss": 1.9704641103744507,
|
277 |
+
"eval_runtime": 11963.7945,
|
278 |
+
"eval_samples_per_second": 1.117,
|
279 |
+
"eval_steps_per_second": 0.14,
|
280 |
+
"step": 800
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 0.27,
|
284 |
+
"learning_rate": 0.0001647214854111406,
|
285 |
+
"loss": 1.8032,
|
286 |
+
"step": 820
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 0.28,
|
290 |
+
"learning_rate": 0.00016206896551724137,
|
291 |
+
"loss": 1.8709,
|
292 |
+
"step": 840
|
293 |
+
},
|
294 |
+
{
|
295 |
+
"epoch": 0.29,
|
296 |
+
"learning_rate": 0.00015941644562334216,
|
297 |
+
"loss": 1.8801,
|
298 |
+
"step": 860
|
299 |
+
},
|
300 |
+
{
|
301 |
+
"epoch": 0.29,
|
302 |
+
"learning_rate": 0.00015676392572944298,
|
303 |
+
"loss": 1.8535,
|
304 |
+
"step": 880
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 0.3,
|
308 |
+
"learning_rate": 0.00015411140583554373,
|
309 |
+
"loss": 1.8631,
|
310 |
+
"step": 900
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.31,
|
314 |
+
"learning_rate": 0.00015145888594164455,
|
315 |
+
"loss": 1.752,
|
316 |
+
"step": 920
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 0.31,
|
320 |
+
"learning_rate": 0.00014880636604774534,
|
321 |
+
"loss": 1.8705,
|
322 |
+
"step": 940
|
323 |
+
},
|
324 |
+
{
|
325 |
+
"epoch": 0.32,
|
326 |
+
"learning_rate": 0.00014615384615384615,
|
327 |
+
"loss": 1.8301,
|
328 |
+
"step": 960
|
329 |
+
},
|
330 |
+
{
|
331 |
+
"epoch": 0.33,
|
332 |
+
"learning_rate": 0.00014350132625994694,
|
333 |
+
"loss": 1.7892,
|
334 |
+
"step": 980
|
335 |
+
},
|
336 |
+
{
|
337 |
+
"epoch": 0.33,
|
338 |
+
"learning_rate": 0.00014084880636604772,
|
339 |
+
"loss": 1.8671,
|
340 |
+
"step": 1000
|
341 |
+
},
|
342 |
+
{
|
343 |
+
"epoch": 0.33,
|
344 |
+
"eval_loss": 1.9677250385284424,
|
345 |
+
"eval_runtime": 11951.074,
|
346 |
+
"eval_samples_per_second": 1.119,
|
347 |
+
"eval_steps_per_second": 0.14,
|
348 |
+
"step": 1000
|
349 |
+
},
|
350 |
+
{
|
351 |
+
"epoch": 0.34,
|
352 |
+
"learning_rate": 0.00013819628647214854,
|
353 |
+
"loss": 1.8168,
|
354 |
+
"step": 1020
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 0.34,
|
358 |
+
"learning_rate": 0.00013554376657824932,
|
359 |
+
"loss": 1.8268,
|
360 |
+
"step": 1040
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 0.35,
|
364 |
+
"learning_rate": 0.00013289124668435014,
|
365 |
+
"loss": 1.8073,
|
366 |
+
"step": 1060
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.36,
|
370 |
+
"learning_rate": 0.0001302387267904509,
|
371 |
+
"loss": 1.9313,
|
372 |
+
"step": 1080
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 0.36,
|
376 |
+
"learning_rate": 0.0001275862068965517,
|
377 |
+
"loss": 1.9097,
|
378 |
+
"step": 1100
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 0.37,
|
382 |
+
"learning_rate": 0.0001249336870026525,
|
383 |
+
"loss": 1.9032,
|
384 |
+
"step": 1120
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 0.38,
|
388 |
+
"learning_rate": 0.0001222811671087533,
|
389 |
+
"loss": 1.8207,
|
390 |
+
"step": 1140
|
391 |
+
},
|
392 |
+
{
|
393 |
+
"epoch": 0.38,
|
394 |
+
"learning_rate": 0.0001196286472148541,
|
395 |
+
"loss": 1.8815,
|
396 |
+
"step": 1160
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 0.39,
|
400 |
+
"learning_rate": 0.0001169761273209549,
|
401 |
+
"loss": 1.8297,
|
402 |
+
"step": 1180
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 0.4,
|
406 |
+
"learning_rate": 0.0001143236074270557,
|
407 |
+
"loss": 1.8899,
|
408 |
+
"step": 1200
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.4,
|
412 |
+
"eval_loss": 1.9722812175750732,
|
413 |
+
"eval_runtime": 11951.1639,
|
414 |
+
"eval_samples_per_second": 1.119,
|
415 |
+
"eval_steps_per_second": 0.14,
|
416 |
+
"step": 1200
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 0.4,
|
420 |
+
"learning_rate": 0.0001116710875331565,
|
421 |
+
"loss": 1.8803,
|
422 |
+
"step": 1220
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.41,
|
426 |
+
"learning_rate": 0.00010901856763925729,
|
427 |
+
"loss": 1.8548,
|
428 |
+
"step": 1240
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 0.42,
|
432 |
+
"learning_rate": 0.00010636604774535807,
|
433 |
+
"loss": 1.8498,
|
434 |
+
"step": 1260
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 0.42,
|
438 |
+
"learning_rate": 0.00010371352785145887,
|
439 |
+
"loss": 1.8057,
|
440 |
+
"step": 1280
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 0.43,
|
444 |
+
"learning_rate": 0.00010106100795755967,
|
445 |
+
"loss": 1.8572,
|
446 |
+
"step": 1300
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 0.44,
|
450 |
+
"learning_rate": 9.840848806366047e-05,
|
451 |
+
"loss": 1.8469,
|
452 |
+
"step": 1320
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 0.44,
|
456 |
+
"learning_rate": 9.575596816976126e-05,
|
457 |
+
"loss": 1.8405,
|
458 |
+
"step": 1340
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 0.45,
|
462 |
+
"learning_rate": 9.310344827586206e-05,
|
463 |
+
"loss": 1.8193,
|
464 |
+
"step": 1360
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.46,
|
468 |
+
"learning_rate": 9.045092838196286e-05,
|
469 |
+
"loss": 1.7731,
|
470 |
+
"step": 1380
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 0.46,
|
474 |
+
"learning_rate": 8.779840848806366e-05,
|
475 |
+
"loss": 1.8329,
|
476 |
+
"step": 1400
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 0.46,
|
480 |
+
"eval_loss": 1.9736511707305908,
|
481 |
+
"eval_runtime": 11956.5916,
|
482 |
+
"eval_samples_per_second": 1.118,
|
483 |
+
"eval_steps_per_second": 0.14,
|
484 |
+
"step": 1400
|
485 |
+
},
|
486 |
+
{
|
487 |
+
"epoch": 0.47,
|
488 |
+
"learning_rate": 8.514588859416444e-05,
|
489 |
+
"loss": 1.8872,
|
490 |
+
"step": 1420
|
491 |
+
},
|
492 |
+
{
|
493 |
+
"epoch": 0.48,
|
494 |
+
"learning_rate": 8.249336870026524e-05,
|
495 |
+
"loss": 1.8511,
|
496 |
+
"step": 1440
|
497 |
+
},
|
498 |
+
{
|
499 |
+
"epoch": 0.48,
|
500 |
+
"learning_rate": 7.984084880636604e-05,
|
501 |
+
"loss": 1.7523,
|
502 |
+
"step": 1460
|
503 |
+
},
|
504 |
+
{
|
505 |
+
"epoch": 0.49,
|
506 |
+
"learning_rate": 7.718832891246684e-05,
|
507 |
+
"loss": 1.8619,
|
508 |
+
"step": 1480
|
509 |
+
},
|
510 |
+
{
|
511 |
+
"epoch": 0.5,
|
512 |
+
"learning_rate": 7.453580901856764e-05,
|
513 |
+
"loss": 1.8363,
|
514 |
+
"step": 1500
|
515 |
+
},
|
516 |
+
{
|
517 |
+
"epoch": 0.5,
|
518 |
+
"learning_rate": 7.188328912466844e-05,
|
519 |
+
"loss": 1.7745,
|
520 |
+
"step": 1520
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.51,
|
524 |
+
"learning_rate": 6.923076923076922e-05,
|
525 |
+
"loss": 1.8164,
|
526 |
+
"step": 1540
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 0.52,
|
530 |
+
"learning_rate": 6.657824933687002e-05,
|
531 |
+
"loss": 1.7997,
|
532 |
+
"step": 1560
|
533 |
+
},
|
534 |
+
{
|
535 |
+
"epoch": 0.52,
|
536 |
+
"learning_rate": 6.392572944297081e-05,
|
537 |
+
"loss": 1.8377,
|
538 |
+
"step": 1580
|
539 |
+
},
|
540 |
+
{
|
541 |
+
"epoch": 0.53,
|
542 |
+
"learning_rate": 6.127320954907161e-05,
|
543 |
+
"loss": 1.851,
|
544 |
+
"step": 1600
|
545 |
+
},
|
546 |
+
{
|
547 |
+
"epoch": 0.53,
|
548 |
+
"eval_loss": 1.9729183912277222,
|
549 |
+
"eval_runtime": 11970.5767,
|
550 |
+
"eval_samples_per_second": 1.117,
|
551 |
+
"eval_steps_per_second": 0.14,
|
552 |
+
"step": 1600
|
553 |
+
},
|
554 |
+
{
|
555 |
+
"epoch": 0.54,
|
556 |
+
"learning_rate": 5.862068965517241e-05,
|
557 |
+
"loss": 1.8382,
|
558 |
+
"step": 1620
|
559 |
+
},
|
560 |
+
{
|
561 |
+
"epoch": 0.54,
|
562 |
+
"learning_rate": 5.59681697612732e-05,
|
563 |
+
"loss": 1.8863,
|
564 |
+
"step": 1640
|
565 |
+
},
|
566 |
+
{
|
567 |
+
"epoch": 0.55,
|
568 |
+
"learning_rate": 5.3315649867374e-05,
|
569 |
+
"loss": 1.762,
|
570 |
+
"step": 1660
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 0.56,
|
574 |
+
"learning_rate": 5.06631299734748e-05,
|
575 |
+
"loss": 1.822,
|
576 |
+
"step": 1680
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.56,
|
580 |
+
"learning_rate": 4.801061007957559e-05,
|
581 |
+
"loss": 1.7776,
|
582 |
+
"step": 1700
|
583 |
+
},
|
584 |
+
{
|
585 |
+
"epoch": 0.57,
|
586 |
+
"learning_rate": 4.5358090185676386e-05,
|
587 |
+
"loss": 1.809,
|
588 |
+
"step": 1720
|
589 |
+
},
|
590 |
+
{
|
591 |
+
"epoch": 0.58,
|
592 |
+
"learning_rate": 4.2705570291777186e-05,
|
593 |
+
"loss": 1.912,
|
594 |
+
"step": 1740
|
595 |
+
},
|
596 |
+
{
|
597 |
+
"epoch": 0.58,
|
598 |
+
"learning_rate": 4.005305039787798e-05,
|
599 |
+
"loss": 1.7557,
|
600 |
+
"step": 1760
|
601 |
+
},
|
602 |
+
{
|
603 |
+
"epoch": 0.59,
|
604 |
+
"learning_rate": 3.7400530503978774e-05,
|
605 |
+
"loss": 1.8232,
|
606 |
+
"step": 1780
|
607 |
+
},
|
608 |
+
{
|
609 |
+
"epoch": 0.6,
|
610 |
+
"learning_rate": 3.4748010610079574e-05,
|
611 |
+
"loss": 1.8613,
|
612 |
+
"step": 1800
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 0.6,
|
616 |
+
"eval_loss": 1.9704504013061523,
|
617 |
+
"eval_runtime": 11980.5334,
|
618 |
+
"eval_samples_per_second": 1.116,
|
619 |
+
"eval_steps_per_second": 0.139,
|
620 |
+
"step": 1800
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 0.6,
|
624 |
+
"learning_rate": 3.209549071618037e-05,
|
625 |
+
"loss": 1.8378,
|
626 |
+
"step": 1820
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 0.61,
|
630 |
+
"learning_rate": 2.9442970822281164e-05,
|
631 |
+
"loss": 1.8871,
|
632 |
+
"step": 1840
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.62,
|
636 |
+
"learning_rate": 2.679045092838196e-05,
|
637 |
+
"loss": 1.8821,
|
638 |
+
"step": 1860
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 0.62,
|
642 |
+
"learning_rate": 2.4137931034482755e-05,
|
643 |
+
"loss": 1.8487,
|
644 |
+
"step": 1880
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 0.63,
|
648 |
+
"learning_rate": 2.1485411140583555e-05,
|
649 |
+
"loss": 1.8439,
|
650 |
+
"step": 1900
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 0.64,
|
654 |
+
"learning_rate": 1.883289124668435e-05,
|
655 |
+
"loss": 1.8469,
|
656 |
+
"step": 1920
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 0.64,
|
660 |
+
"learning_rate": 1.6180371352785142e-05,
|
661 |
+
"loss": 1.7797,
|
662 |
+
"step": 1940
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 0.65,
|
666 |
+
"learning_rate": 1.352785145888594e-05,
|
667 |
+
"loss": 1.7795,
|
668 |
+
"step": 1960
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 0.66,
|
672 |
+
"learning_rate": 1.0875331564986736e-05,
|
673 |
+
"loss": 1.8483,
|
674 |
+
"step": 1980
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.66,
|
678 |
+
"learning_rate": 8.222811671087533e-06,
|
679 |
+
"loss": 1.8321,
|
680 |
+
"step": 2000
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 0.66,
|
684 |
+
"eval_loss": 1.9704707860946655,
|
685 |
+
"eval_runtime": 11964.4776,
|
686 |
+
"eval_samples_per_second": 1.117,
|
687 |
+
"eval_steps_per_second": 0.14,
|
688 |
+
"step": 2000
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.67,
|
692 |
+
"learning_rate": 5.5702917771883284e-06,
|
693 |
+
"loss": 1.7708,
|
694 |
+
"step": 2020
|
695 |
+
},
|
696 |
+
{
|
697 |
+
"epoch": 0.68,
|
698 |
+
"learning_rate": 2.9177718832891245e-06,
|
699 |
+
"loss": 1.8731,
|
700 |
+
"step": 2040
|
701 |
+
},
|
702 |
+
{
|
703 |
+
"epoch": 0.68,
|
704 |
+
"learning_rate": 2.652519893899204e-07,
|
705 |
+
"loss": 1.8181,
|
706 |
+
"step": 2060
|
707 |
+
},
|
708 |
+
{
|
709 |
+
"epoch": 0.69,
|
710 |
+
"learning_rate": 0.0,
|
711 |
+
"loss": 1.8692,
|
712 |
+
"step": 2080
|
713 |
+
},
|
714 |
+
{
|
715 |
+
"epoch": 0.7,
|
716 |
+
"learning_rate": 0.0,
|
717 |
+
"loss": 1.8602,
|
718 |
+
"step": 2100
|
719 |
+
},
|
720 |
+
{
|
721 |
+
"epoch": 0.7,
|
722 |
+
"learning_rate": 0.0,
|
723 |
+
"loss": 1.8891,
|
724 |
+
"step": 2120
|
725 |
+
},
|
726 |
+
{
|
727 |
+
"epoch": 0.71,
|
728 |
+
"learning_rate": 0.0,
|
729 |
+
"loss": 1.8205,
|
730 |
+
"step": 2140
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.72,
|
734 |
+
"learning_rate": 0.0,
|
735 |
+
"loss": 1.8242,
|
736 |
+
"step": 2160
|
737 |
+
},
|
738 |
+
{
|
739 |
+
"epoch": 0.72,
|
740 |
+
"learning_rate": 0.0,
|
741 |
+
"loss": 1.8661,
|
742 |
+
"step": 2180
|
743 |
+
},
|
744 |
+
{
|
745 |
+
"epoch": 0.73,
|
746 |
+
"learning_rate": 0.0,
|
747 |
+
"loss": 1.7982,
|
748 |
+
"step": 2200
|
749 |
+
},
|
750 |
+
{
|
751 |
+
"epoch": 0.73,
|
752 |
+
"eval_loss": 1.970503568649292,
|
753 |
+
"eval_runtime": 11950.1314,
|
754 |
+
"eval_samples_per_second": 1.119,
|
755 |
+
"eval_steps_per_second": 0.14,
|
756 |
+
"step": 2200
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 0.74,
|
760 |
+
"learning_rate": 0.0,
|
761 |
+
"loss": 1.7967,
|
762 |
+
"step": 2220
|
763 |
+
},
|
764 |
+
{
|
765 |
+
"epoch": 0.74,
|
766 |
+
"learning_rate": 0.0,
|
767 |
+
"loss": 1.8346,
|
768 |
+
"step": 2240
|
769 |
+
},
|
770 |
+
{
|
771 |
+
"epoch": 0.75,
|
772 |
+
"learning_rate": 0.0,
|
773 |
+
"loss": 1.8242,
|
774 |
+
"step": 2260
|
775 |
+
},
|
776 |
+
{
|
777 |
+
"epoch": 0.76,
|
778 |
+
"learning_rate": 0.0,
|
779 |
+
"loss": 1.8429,
|
780 |
+
"step": 2280
|
781 |
+
},
|
782 |
+
{
|
783 |
+
"epoch": 0.76,
|
784 |
+
"learning_rate": 0.0,
|
785 |
+
"loss": 1.8795,
|
786 |
+
"step": 2300
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.77,
|
790 |
+
"learning_rate": 0.0,
|
791 |
+
"loss": 1.8967,
|
792 |
+
"step": 2320
|
793 |
+
},
|
794 |
+
{
|
795 |
+
"epoch": 0.78,
|
796 |
+
"learning_rate": 0.0,
|
797 |
+
"loss": 1.815,
|
798 |
+
"step": 2340
|
799 |
+
},
|
800 |
+
{
|
801 |
+
"epoch": 0.78,
|
802 |
+
"learning_rate": 0.0,
|
803 |
+
"loss": 1.8165,
|
804 |
+
"step": 2360
|
805 |
+
},
|
806 |
+
{
|
807 |
+
"epoch": 0.79,
|
808 |
+
"learning_rate": 0.0,
|
809 |
+
"loss": 1.838,
|
810 |
+
"step": 2380
|
811 |
+
},
|
812 |
+
{
|
813 |
+
"epoch": 0.8,
|
814 |
+
"learning_rate": 0.0,
|
815 |
+
"loss": 1.7798,
|
816 |
+
"step": 2400
|
817 |
+
},
|
818 |
+
{
|
819 |
+
"epoch": 0.8,
|
820 |
+
"eval_loss": 1.970503568649292,
|
821 |
+
"eval_runtime": 11951.0888,
|
822 |
+
"eval_samples_per_second": 1.119,
|
823 |
+
"eval_steps_per_second": 0.14,
|
824 |
+
"step": 2400
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 0.8,
|
828 |
+
"learning_rate": 0.0,
|
829 |
+
"loss": 1.7999,
|
830 |
+
"step": 2420
|
831 |
+
},
|
832 |
+
{
|
833 |
+
"epoch": 0.81,
|
834 |
+
"learning_rate": 0.0,
|
835 |
+
"loss": 1.7957,
|
836 |
+
"step": 2440
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 0.82,
|
840 |
+
"learning_rate": 0.0,
|
841 |
+
"loss": 1.8306,
|
842 |
+
"step": 2460
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.82,
|
846 |
+
"learning_rate": 0.0,
|
847 |
+
"loss": 1.87,
|
848 |
+
"step": 2480
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 0.83,
|
852 |
+
"learning_rate": 0.0,
|
853 |
+
"loss": 1.8046,
|
854 |
+
"step": 2500
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 0.84,
|
858 |
+
"learning_rate": 0.0,
|
859 |
+
"loss": 1.76,
|
860 |
+
"step": 2520
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 0.84,
|
864 |
+
"learning_rate": 0.0,
|
865 |
+
"loss": 1.869,
|
866 |
+
"step": 2540
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 0.85,
|
870 |
+
"learning_rate": 0.0,
|
871 |
+
"loss": 1.8684,
|
872 |
+
"step": 2560
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 0.86,
|
876 |
+
"learning_rate": 0.0,
|
877 |
+
"loss": 1.8757,
|
878 |
+
"step": 2580
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 0.86,
|
882 |
+
"learning_rate": 0.0,
|
883 |
+
"loss": 1.868,
|
884 |
+
"step": 2600
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.86,
|
888 |
+
"eval_loss": 1.970503568649292,
|
889 |
+
"eval_runtime": 11952.488,
|
890 |
+
"eval_samples_per_second": 1.118,
|
891 |
+
"eval_steps_per_second": 0.14,
|
892 |
+
"step": 2600
|
893 |
+
},
|
894 |
+
{
|
895 |
+
"epoch": 0.87,
|
896 |
+
"learning_rate": 0.0,
|
897 |
+
"loss": 1.8228,
|
898 |
+
"step": 2620
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.88,
|
902 |
+
"learning_rate": 0.0,
|
903 |
+
"loss": 1.8721,
|
904 |
+
"step": 2640
|
905 |
+
},
|
906 |
+
{
|
907 |
+
"epoch": 0.88,
|
908 |
+
"learning_rate": 0.0,
|
909 |
+
"loss": 1.8557,
|
910 |
+
"step": 2660
|
911 |
+
},
|
912 |
+
{
|
913 |
+
"epoch": 0.89,
|
914 |
+
"learning_rate": 0.0,
|
915 |
+
"loss": 1.8545,
|
916 |
+
"step": 2680
|
917 |
+
},
|
918 |
+
{
|
919 |
+
"epoch": 0.9,
|
920 |
+
"learning_rate": 0.0,
|
921 |
+
"loss": 1.8366,
|
922 |
+
"step": 2700
|
923 |
+
},
|
924 |
+
{
|
925 |
+
"epoch": 0.9,
|
926 |
+
"learning_rate": 0.0,
|
927 |
+
"loss": 1.8381,
|
928 |
+
"step": 2720
|
929 |
+
},
|
930 |
+
{
|
931 |
+
"epoch": 0.91,
|
932 |
+
"learning_rate": 0.0,
|
933 |
+
"loss": 1.8305,
|
934 |
+
"step": 2740
|
935 |
+
},
|
936 |
+
{
|
937 |
+
"epoch": 0.92,
|
938 |
+
"learning_rate": 0.0,
|
939 |
+
"loss": 1.8162,
|
940 |
+
"step": 2760
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.92,
|
944 |
+
"learning_rate": 0.0,
|
945 |
+
"loss": 1.8109,
|
946 |
+
"step": 2780
|
947 |
+
},
|
948 |
+
{
|
949 |
+
"epoch": 0.93,
|
950 |
+
"learning_rate": 0.0,
|
951 |
+
"loss": 1.8694,
|
952 |
+
"step": 2800
|
953 |
+
},
|
954 |
+
{
|
955 |
+
"epoch": 0.93,
|
956 |
+
"eval_loss": 1.970503568649292,
|
957 |
+
"eval_runtime": 11969.9504,
|
958 |
+
"eval_samples_per_second": 1.117,
|
959 |
+
"eval_steps_per_second": 0.14,
|
960 |
+
"step": 2800
|
961 |
+
},
|
962 |
+
{
|
963 |
+
"epoch": 0.94,
|
964 |
+
"learning_rate": 0.0,
|
965 |
+
"loss": 1.8201,
|
966 |
+
"step": 2820
|
967 |
+
},
|
968 |
+
{
|
969 |
+
"epoch": 0.94,
|
970 |
+
"learning_rate": 0.0,
|
971 |
+
"loss": 1.848,
|
972 |
+
"step": 2840
|
973 |
+
},
|
974 |
+
{
|
975 |
+
"epoch": 0.95,
|
976 |
+
"learning_rate": 0.0,
|
977 |
+
"loss": 1.8138,
|
978 |
+
"step": 2860
|
979 |
+
},
|
980 |
+
{
|
981 |
+
"epoch": 0.96,
|
982 |
+
"learning_rate": 0.0,
|
983 |
+
"loss": 1.8554,
|
984 |
+
"step": 2880
|
985 |
+
},
|
986 |
+
{
|
987 |
+
"epoch": 0.96,
|
988 |
+
"learning_rate": 0.0,
|
989 |
+
"loss": 1.8321,
|
990 |
+
"step": 2900
|
991 |
+
},
|
992 |
+
{
|
993 |
+
"epoch": 0.97,
|
994 |
+
"learning_rate": 0.0,
|
995 |
+
"loss": 1.8404,
|
996 |
+
"step": 2920
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.98,
|
1000 |
+
"learning_rate": 0.0,
|
1001 |
+
"loss": 1.7883,
|
1002 |
+
"step": 2940
|
1003 |
+
},
|
1004 |
+
{
|
1005 |
+
"epoch": 0.98,
|
1006 |
+
"learning_rate": 0.0,
|
1007 |
+
"loss": 1.8733,
|
1008 |
+
"step": 2960
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 0.99,
|
1012 |
+
"learning_rate": 0.0,
|
1013 |
+
"loss": 1.8074,
|
1014 |
+
"step": 2980
|
1015 |
+
},
|
1016 |
+
{
|
1017 |
+
"epoch": 1.0,
|
1018 |
+
"learning_rate": 0.0,
|
1019 |
+
"loss": 1.887,
|
1020 |
+
"step": 3000
|
1021 |
+
},
|
1022 |
+
{
|
1023 |
+
"epoch": 1.0,
|
1024 |
+
"eval_loss": 1.970503568649292,
|
1025 |
+
"eval_runtime": 11974.5637,
|
1026 |
+
"eval_samples_per_second": 1.116,
|
1027 |
+
"eval_steps_per_second": 0.14,
|
1028 |
+
"step": 3000
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 1.0,
|
1032 |
+
"learning_rate": 0.0,
|
1033 |
+
"loss": 1.7165,
|
1034 |
+
"step": 3020
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 1.01,
|
1038 |
+
"learning_rate": 0.0,
|
1039 |
+
"loss": 1.8041,
|
1040 |
+
"step": 3040
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 1.01,
|
1044 |
+
"learning_rate": 0.0,
|
1045 |
+
"loss": 1.8031,
|
1046 |
+
"step": 3060
|
1047 |
+
},
|
1048 |
+
{
|
1049 |
+
"epoch": 1.02,
|
1050 |
+
"learning_rate": 0.0,
|
1051 |
+
"loss": 1.8493,
|
1052 |
+
"step": 3080
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 1.03,
|
1056 |
+
"learning_rate": 0.0,
|
1057 |
+
"loss": 1.8475,
|
1058 |
+
"step": 3100
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 1.03,
|
1062 |
+
"learning_rate": 0.0,
|
1063 |
+
"loss": 1.7994,
|
1064 |
+
"step": 3120
|
1065 |
+
},
|
1066 |
+
{
|
1067 |
+
"epoch": 1.04,
|
1068 |
+
"learning_rate": 0.0,
|
1069 |
+
"loss": 1.7904,
|
1070 |
+
"step": 3140
|
1071 |
+
},
|
1072 |
+
{
|
1073 |
+
"epoch": 1.05,
|
1074 |
+
"learning_rate": 0.0,
|
1075 |
+
"loss": 1.8078,
|
1076 |
+
"step": 3160
|
1077 |
+
},
|
1078 |
+
{
|
1079 |
+
"epoch": 1.05,
|
1080 |
+
"learning_rate": 0.0,
|
1081 |
+
"loss": 1.8625,
|
1082 |
+
"step": 3180
|
1083 |
+
},
|
1084 |
+
{
|
1085 |
+
"epoch": 1.06,
|
1086 |
+
"learning_rate": 0.0,
|
1087 |
+
"loss": 1.7874,
|
1088 |
+
"step": 3200
|
1089 |
+
},
|
1090 |
+
{
|
1091 |
+
"epoch": 1.06,
|
1092 |
+
"eval_loss": 1.970503568649292,
|
1093 |
+
"eval_runtime": 11966.566,
|
1094 |
+
"eval_samples_per_second": 1.117,
|
1095 |
+
"eval_steps_per_second": 0.14,
|
1096 |
+
"step": 3200
|
1097 |
+
},
|
1098 |
+
{
|
1099 |
+
"epoch": 1.07,
|
1100 |
+
"learning_rate": 0.0,
|
1101 |
+
"loss": 1.7956,
|
1102 |
+
"step": 3220
|
1103 |
+
},
|
1104 |
+
{
|
1105 |
+
"epoch": 1.07,
|
1106 |
+
"learning_rate": 0.0,
|
1107 |
+
"loss": 1.7517,
|
1108 |
+
"step": 3240
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 1.08,
|
1112 |
+
"learning_rate": 0.0,
|
1113 |
+
"loss": 1.7861,
|
1114 |
+
"step": 3260
|
1115 |
+
},
|
1116 |
+
{
|
1117 |
+
"epoch": 1.09,
|
1118 |
+
"learning_rate": 0.0,
|
1119 |
+
"loss": 1.7957,
|
1120 |
+
"step": 3280
|
1121 |
+
},
|
1122 |
+
{
|
1123 |
+
"epoch": 1.09,
|
1124 |
+
"learning_rate": 0.0,
|
1125 |
+
"loss": 1.8147,
|
1126 |
+
"step": 3300
|
1127 |
+
},
|
1128 |
+
{
|
1129 |
+
"epoch": 1.1,
|
1130 |
+
"learning_rate": 0.0,
|
1131 |
+
"loss": 1.7795,
|
1132 |
+
"step": 3320
|
1133 |
+
},
|
1134 |
+
{
|
1135 |
+
"epoch": 1.11,
|
1136 |
+
"learning_rate": 0.0,
|
1137 |
+
"loss": 1.7532,
|
1138 |
+
"step": 3340
|
1139 |
+
},
|
1140 |
+
{
|
1141 |
+
"epoch": 1.11,
|
1142 |
+
"learning_rate": 0.0,
|
1143 |
+
"loss": 1.8414,
|
1144 |
+
"step": 3360
|
1145 |
+
},
|
1146 |
+
{
|
1147 |
+
"epoch": 1.12,
|
1148 |
+
"learning_rate": 0.0,
|
1149 |
+
"loss": 1.771,
|
1150 |
+
"step": 3380
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 1.13,
|
1154 |
+
"learning_rate": 0.0,
|
1155 |
+
"loss": 1.8119,
|
1156 |
+
"step": 3400
|
1157 |
+
},
|
1158 |
+
{
|
1159 |
+
"epoch": 1.13,
|
1160 |
+
"eval_loss": 1.970503568649292,
|
1161 |
+
"eval_runtime": 11954.7549,
|
1162 |
+
"eval_samples_per_second": 1.118,
|
1163 |
+
"eval_steps_per_second": 0.14,
|
1164 |
+
"step": 3400
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 1.13,
|
1168 |
+
"learning_rate": 0.0,
|
1169 |
+
"loss": 1.7944,
|
1170 |
+
"step": 3420
|
1171 |
+
},
|
1172 |
+
{
|
1173 |
+
"epoch": 1.14,
|
1174 |
+
"learning_rate": 0.0,
|
1175 |
+
"loss": 1.7855,
|
1176 |
+
"step": 3440
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 1.15,
|
1180 |
+
"learning_rate": 0.0,
|
1181 |
+
"loss": 1.8574,
|
1182 |
+
"step": 3460
|
1183 |
+
},
|
1184 |
+
{
|
1185 |
+
"epoch": 1.15,
|
1186 |
+
"learning_rate": 0.0,
|
1187 |
+
"loss": 1.8189,
|
1188 |
+
"step": 3480
|
1189 |
+
},
|
1190 |
+
{
|
1191 |
+
"epoch": 1.16,
|
1192 |
+
"learning_rate": 0.0,
|
1193 |
+
"loss": 1.8033,
|
1194 |
+
"step": 3500
|
1195 |
+
},
|
1196 |
+
{
|
1197 |
+
"epoch": 1.17,
|
1198 |
+
"learning_rate": 0.0,
|
1199 |
+
"loss": 1.7689,
|
1200 |
+
"step": 3520
|
1201 |
+
},
|
1202 |
+
{
|
1203 |
+
"epoch": 1.17,
|
1204 |
+
"learning_rate": 0.0,
|
1205 |
+
"loss": 1.8455,
|
1206 |
+
"step": 3540
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 1.18,
|
1210 |
+
"learning_rate": 0.0,
|
1211 |
+
"loss": 1.8255,
|
1212 |
+
"step": 3560
|
1213 |
+
},
|
1214 |
+
{
|
1215 |
+
"epoch": 1.19,
|
1216 |
+
"learning_rate": 0.0,
|
1217 |
+
"loss": 1.8431,
|
1218 |
+
"step": 3580
|
1219 |
+
},
|
1220 |
+
{
|
1221 |
+
"epoch": 1.19,
|
1222 |
+
"learning_rate": 0.0,
|
1223 |
+
"loss": 1.7924,
|
1224 |
+
"step": 3600
|
1225 |
+
},
|
1226 |
+
{
|
1227 |
+
"epoch": 1.19,
|
1228 |
+
"eval_loss": 1.970503568649292,
|
1229 |
+
"eval_runtime": 11953.5885,
|
1230 |
+
"eval_samples_per_second": 1.118,
|
1231 |
+
"eval_steps_per_second": 0.14,
|
1232 |
+
"step": 3600
|
1233 |
+
},
|
1234 |
+
{
|
1235 |
+
"epoch": 1.2,
|
1236 |
+
"learning_rate": 0.0,
|
1237 |
+
"loss": 1.8475,
|
1238 |
+
"step": 3620
|
1239 |
+
},
|
1240 |
+
{
|
1241 |
+
"epoch": 1.21,
|
1242 |
+
"learning_rate": 0.0,
|
1243 |
+
"loss": 1.6746,
|
1244 |
+
"step": 3640
|
1245 |
+
},
|
1246 |
+
{
|
1247 |
+
"epoch": 1.21,
|
1248 |
+
"learning_rate": 0.0,
|
1249 |
+
"loss": 1.8006,
|
1250 |
+
"step": 3660
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 1.22,
|
1254 |
+
"learning_rate": 0.0,
|
1255 |
+
"loss": 1.816,
|
1256 |
+
"step": 3680
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 1.23,
|
1260 |
+
"learning_rate": 0.0,
|
1261 |
+
"loss": 1.7486,
|
1262 |
+
"step": 3700
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 1.23,
|
1266 |
+
"learning_rate": 0.0,
|
1267 |
+
"loss": 1.8086,
|
1268 |
+
"step": 3720
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 1.24,
|
1272 |
+
"learning_rate": 0.0,
|
1273 |
+
"loss": 1.8241,
|
1274 |
+
"step": 3740
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 1.25,
|
1278 |
+
"learning_rate": 0.0,
|
1279 |
+
"loss": 1.823,
|
1280 |
+
"step": 3760
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 1.25,
|
1284 |
+
"learning_rate": 0.0,
|
1285 |
+
"loss": 1.7365,
|
1286 |
+
"step": 3780
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 1.26,
|
1290 |
+
"learning_rate": 0.0,
|
1291 |
+
"loss": 1.801,
|
1292 |
+
"step": 3800
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 1.26,
|
1296 |
+
"eval_loss": 1.970503568649292,
|
1297 |
+
"eval_runtime": 11952.013,
|
1298 |
+
"eval_samples_per_second": 1.118,
|
1299 |
+
"eval_steps_per_second": 0.14,
|
1300 |
+
"step": 3800
|
1301 |
+
},
|
1302 |
+
{
|
1303 |
+
"epoch": 1.27,
|
1304 |
+
"learning_rate": 0.0,
|
1305 |
+
"loss": 1.8497,
|
1306 |
+
"step": 3820
|
1307 |
+
},
|
1308 |
+
{
|
1309 |
+
"epoch": 1.27,
|
1310 |
+
"learning_rate": 0.0,
|
1311 |
+
"loss": 1.7803,
|
1312 |
+
"step": 3840
|
1313 |
+
},
|
1314 |
+
{
|
1315 |
+
"epoch": 1.28,
|
1316 |
+
"learning_rate": 0.0,
|
1317 |
+
"loss": 1.6921,
|
1318 |
+
"step": 3860
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 1.29,
|
1322 |
+
"learning_rate": 0.0,
|
1323 |
+
"loss": 1.8508,
|
1324 |
+
"step": 3880
|
1325 |
+
},
|
1326 |
+
{
|
1327 |
+
"epoch": 1.29,
|
1328 |
+
"learning_rate": 0.0,
|
1329 |
+
"loss": 1.8816,
|
1330 |
+
"step": 3900
|
1331 |
+
},
|
1332 |
+
{
|
1333 |
+
"epoch": 1.3,
|
1334 |
+
"learning_rate": 0.0,
|
1335 |
+
"loss": 1.8105,
|
1336 |
+
"step": 3920
|
1337 |
+
},
|
1338 |
+
{
|
1339 |
+
"epoch": 1.31,
|
1340 |
+
"learning_rate": 0.0,
|
1341 |
+
"loss": 1.8477,
|
1342 |
+
"step": 3940
|
1343 |
+
},
|
1344 |
+
{
|
1345 |
+
"epoch": 1.31,
|
1346 |
+
"learning_rate": 0.0,
|
1347 |
+
"loss": 1.841,
|
1348 |
+
"step": 3960
|
1349 |
+
},
|
1350 |
+
{
|
1351 |
+
"epoch": 1.32,
|
1352 |
+
"learning_rate": 0.0,
|
1353 |
+
"loss": 1.7599,
|
1354 |
+
"step": 3980
|
1355 |
+
},
|
1356 |
+
{
|
1357 |
+
"epoch": 1.33,
|
1358 |
+
"learning_rate": 0.0,
|
1359 |
+
"loss": 1.7074,
|
1360 |
+
"step": 4000
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 1.33,
|
1364 |
+
"eval_loss": 1.970503568649292,
|
1365 |
+
"eval_runtime": 11952.3623,
|
1366 |
+
"eval_samples_per_second": 1.118,
|
1367 |
+
"eval_steps_per_second": 0.14,
|
1368 |
+
"step": 4000
|
1369 |
+
},
|
1370 |
+
{
|
1371 |
+
"epoch": 1.33,
|
1372 |
+
"learning_rate": 0.0,
|
1373 |
+
"loss": 1.7529,
|
1374 |
+
"step": 4020
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 1.34,
|
1378 |
+
"learning_rate": 0.0,
|
1379 |
+
"loss": 1.8928,
|
1380 |
+
"step": 4040
|
1381 |
+
},
|
1382 |
+
{
|
1383 |
+
"epoch": 1.35,
|
1384 |
+
"learning_rate": 0.0,
|
1385 |
+
"loss": 1.8585,
|
1386 |
+
"step": 4060
|
1387 |
+
},
|
1388 |
+
{
|
1389 |
+
"epoch": 1.35,
|
1390 |
+
"learning_rate": 0.0,
|
1391 |
+
"loss": 1.8279,
|
1392 |
+
"step": 4080
|
1393 |
+
},
|
1394 |
+
{
|
1395 |
+
"epoch": 1.36,
|
1396 |
+
"learning_rate": 0.0,
|
1397 |
+
"loss": 1.7949,
|
1398 |
+
"step": 4100
|
1399 |
+
},
|
1400 |
+
{
|
1401 |
+
"epoch": 1.37,
|
1402 |
+
"learning_rate": 0.0,
|
1403 |
+
"loss": 1.8488,
|
1404 |
+
"step": 4120
|
1405 |
+
},
|
1406 |
+
{
|
1407 |
+
"epoch": 1.37,
|
1408 |
+
"learning_rate": 0.0,
|
1409 |
+
"loss": 1.7572,
|
1410 |
+
"step": 4140
|
1411 |
+
},
|
1412 |
+
{
|
1413 |
+
"epoch": 1.38,
|
1414 |
+
"learning_rate": 0.0,
|
1415 |
+
"loss": 1.7533,
|
1416 |
+
"step": 4160
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 1.39,
|
1420 |
+
"learning_rate": 0.0,
|
1421 |
+
"loss": 1.772,
|
1422 |
+
"step": 4180
|
1423 |
+
},
|
1424 |
+
{
|
1425 |
+
"epoch": 1.39,
|
1426 |
+
"learning_rate": 0.0,
|
1427 |
+
"loss": 1.8338,
|
1428 |
+
"step": 4200
|
1429 |
+
},
|
1430 |
+
{
|
1431 |
+
"epoch": 1.39,
|
1432 |
+
"eval_loss": 1.970503568649292,
|
1433 |
+
"eval_runtime": 11954.2113,
|
1434 |
+
"eval_samples_per_second": 1.118,
|
1435 |
+
"eval_steps_per_second": 0.14,
|
1436 |
+
"step": 4200
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 1.4,
|
1440 |
+
"learning_rate": 0.0,
|
1441 |
+
"loss": 1.8084,
|
1442 |
+
"step": 4220
|
1443 |
+
},
|
1444 |
+
{
|
1445 |
+
"epoch": 1.41,
|
1446 |
+
"learning_rate": 0.0,
|
1447 |
+
"loss": 1.8791,
|
1448 |
+
"step": 4240
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 1.41,
|
1452 |
+
"learning_rate": 0.0,
|
1453 |
+
"loss": 1.7906,
|
1454 |
+
"step": 4260
|
1455 |
+
},
|
1456 |
+
{
|
1457 |
+
"epoch": 1.42,
|
1458 |
+
"learning_rate": 0.0,
|
1459 |
+
"loss": 1.8669,
|
1460 |
+
"step": 4280
|
1461 |
+
},
|
1462 |
+
{
|
1463 |
+
"epoch": 1.43,
|
1464 |
+
"learning_rate": 0.0,
|
1465 |
+
"loss": 1.8108,
|
1466 |
+
"step": 4300
|
1467 |
+
},
|
1468 |
+
{
|
1469 |
+
"epoch": 1.43,
|
1470 |
+
"learning_rate": 0.0,
|
1471 |
+
"loss": 1.7769,
|
1472 |
+
"step": 4320
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 1.44,
|
1476 |
+
"learning_rate": 0.0,
|
1477 |
+
"loss": 1.7599,
|
1478 |
+
"step": 4340
|
1479 |
+
},
|
1480 |
+
{
|
1481 |
+
"epoch": 1.45,
|
1482 |
+
"learning_rate": 0.0,
|
1483 |
+
"loss": 1.843,
|
1484 |
+
"step": 4360
|
1485 |
+
},
|
1486 |
+
{
|
1487 |
+
"epoch": 1.45,
|
1488 |
+
"learning_rate": 0.0,
|
1489 |
+
"loss": 1.8608,
|
1490 |
+
"step": 4380
|
1491 |
+
},
|
1492 |
+
{
|
1493 |
+
"epoch": 1.46,
|
1494 |
+
"learning_rate": 0.0,
|
1495 |
+
"loss": 1.8382,
|
1496 |
+
"step": 4400
|
1497 |
+
},
|
1498 |
+
{
|
1499 |
+
"epoch": 1.46,
|
1500 |
+
"eval_loss": 1.970503568649292,
|
1501 |
+
"eval_runtime": 11958.7166,
|
1502 |
+
"eval_samples_per_second": 1.118,
|
1503 |
+
"eval_steps_per_second": 0.14,
|
1504 |
+
"step": 4400
|
1505 |
+
},
|
1506 |
+
{
|
1507 |
+
"epoch": 1.47,
|
1508 |
+
"learning_rate": 0.0,
|
1509 |
+
"loss": 1.8002,
|
1510 |
+
"step": 4420
|
1511 |
+
},
|
1512 |
+
{
|
1513 |
+
"epoch": 1.47,
|
1514 |
+
"learning_rate": 0.0,
|
1515 |
+
"loss": 1.8292,
|
1516 |
+
"step": 4440
|
1517 |
+
},
|
1518 |
+
{
|
1519 |
+
"epoch": 1.48,
|
1520 |
+
"learning_rate": 0.0,
|
1521 |
+
"loss": 1.8184,
|
1522 |
+
"step": 4460
|
1523 |
+
},
|
1524 |
+
{
|
1525 |
+
"epoch": 1.49,
|
1526 |
+
"learning_rate": 0.0,
|
1527 |
+
"loss": 1.7712,
|
1528 |
+
"step": 4480
|
1529 |
+
},
|
1530 |
+
{
|
1531 |
+
"epoch": 1.49,
|
1532 |
+
"learning_rate": 0.0,
|
1533 |
+
"loss": 1.8307,
|
1534 |
+
"step": 4500
|
1535 |
+
},
|
1536 |
+
{
|
1537 |
+
"epoch": 1.5,
|
1538 |
+
"learning_rate": 0.0,
|
1539 |
+
"loss": 1.7883,
|
1540 |
+
"step": 4520
|
1541 |
+
},
|
1542 |
+
{
|
1543 |
+
"epoch": 1.51,
|
1544 |
+
"learning_rate": 0.0,
|
1545 |
+
"loss": 1.8106,
|
1546 |
+
"step": 4540
|
1547 |
+
},
|
1548 |
+
{
|
1549 |
+
"epoch": 1.51,
|
1550 |
+
"learning_rate": 0.0,
|
1551 |
+
"loss": 1.7999,
|
1552 |
+
"step": 4560
|
1553 |
+
},
|
1554 |
+
{
|
1555 |
+
"epoch": 1.52,
|
1556 |
+
"learning_rate": 0.0,
|
1557 |
+
"loss": 1.8353,
|
1558 |
+
"step": 4580
|
1559 |
+
},
|
1560 |
+
{
|
1561 |
+
"epoch": 1.53,
|
1562 |
+
"learning_rate": 0.0,
|
1563 |
+
"loss": 1.8366,
|
1564 |
+
"step": 4600
|
1565 |
+
},
|
1566 |
+
{
|
1567 |
+
"epoch": 1.53,
|
1568 |
+
"eval_loss": 1.970503568649292,
|
1569 |
+
"eval_runtime": 11971.0491,
|
1570 |
+
"eval_samples_per_second": 1.117,
|
1571 |
+
"eval_steps_per_second": 0.14,
|
1572 |
+
"step": 4600
|
1573 |
+
},
|
1574 |
+
{
|
1575 |
+
"epoch": 1.53,
|
1576 |
+
"learning_rate": 0.0,
|
1577 |
+
"loss": 1.8018,
|
1578 |
+
"step": 4620
|
1579 |
+
},
|
1580 |
+
{
|
1581 |
+
"epoch": 1.54,
|
1582 |
+
"learning_rate": 0.0,
|
1583 |
+
"loss": 1.84,
|
1584 |
+
"step": 4640
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 1.55,
|
1588 |
+
"learning_rate": 0.0,
|
1589 |
+
"loss": 1.7784,
|
1590 |
+
"step": 4660
|
1591 |
+
},
|
1592 |
+
{
|
1593 |
+
"epoch": 1.55,
|
1594 |
+
"learning_rate": 0.0,
|
1595 |
+
"loss": 1.794,
|
1596 |
+
"step": 4680
|
1597 |
+
},
|
1598 |
+
{
|
1599 |
+
"epoch": 1.56,
|
1600 |
+
"learning_rate": 0.0,
|
1601 |
+
"loss": 1.8237,
|
1602 |
+
"step": 4700
|
1603 |
+
},
|
1604 |
+
{
|
1605 |
+
"epoch": 1.57,
|
1606 |
+
"learning_rate": 0.0,
|
1607 |
+
"loss": 1.7697,
|
1608 |
+
"step": 4720
|
1609 |
+
},
|
1610 |
+
{
|
1611 |
+
"epoch": 1.57,
|
1612 |
+
"learning_rate": 0.0,
|
1613 |
+
"loss": 1.8482,
|
1614 |
+
"step": 4740
|
1615 |
+
},
|
1616 |
+
{
|
1617 |
+
"epoch": 1.58,
|
1618 |
+
"learning_rate": 0.0,
|
1619 |
+
"loss": 1.8008,
|
1620 |
+
"step": 4760
|
1621 |
+
},
|
1622 |
+
{
|
1623 |
+
"epoch": 1.59,
|
1624 |
+
"learning_rate": 0.0,
|
1625 |
+
"loss": 1.8082,
|
1626 |
+
"step": 4780
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 1.59,
|
1630 |
+
"learning_rate": 0.0,
|
1631 |
+
"loss": 1.7799,
|
1632 |
+
"step": 4800
|
1633 |
+
},
|
1634 |
+
{
|
1635 |
+
"epoch": 1.59,
|
1636 |
+
"eval_loss": 1.970503568649292,
|
1637 |
+
"eval_runtime": 11955.9821,
|
1638 |
+
"eval_samples_per_second": 1.118,
|
1639 |
+
"eval_steps_per_second": 0.14,
|
1640 |
+
"step": 4800
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 1.6,
|
1644 |
+
"learning_rate": 0.0,
|
1645 |
+
"loss": 1.8339,
|
1646 |
+
"step": 4820
|
1647 |
+
},
|
1648 |
+
{
|
1649 |
+
"epoch": 1.61,
|
1650 |
+
"learning_rate": 0.0,
|
1651 |
+
"loss": 1.8072,
|
1652 |
+
"step": 4840
|
1653 |
+
},
|
1654 |
+
{
|
1655 |
+
"epoch": 1.61,
|
1656 |
+
"learning_rate": 0.0,
|
1657 |
+
"loss": 1.8024,
|
1658 |
+
"step": 4860
|
1659 |
+
},
|
1660 |
+
{
|
1661 |
+
"epoch": 1.62,
|
1662 |
+
"learning_rate": 0.0,
|
1663 |
+
"loss": 1.8609,
|
1664 |
+
"step": 4880
|
1665 |
+
},
|
1666 |
+
{
|
1667 |
+
"epoch": 1.63,
|
1668 |
+
"learning_rate": 0.0,
|
1669 |
+
"loss": 1.8769,
|
1670 |
+
"step": 4900
|
1671 |
+
},
|
1672 |
+
{
|
1673 |
+
"epoch": 1.63,
|
1674 |
+
"learning_rate": 0.0,
|
1675 |
+
"loss": 1.808,
|
1676 |
+
"step": 4920
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 1.64,
|
1680 |
+
"learning_rate": 0.0,
|
1681 |
+
"loss": 1.8482,
|
1682 |
+
"step": 4940
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 1.65,
|
1686 |
+
"learning_rate": 0.0,
|
1687 |
+
"loss": 1.8116,
|
1688 |
+
"step": 4960
|
1689 |
+
},
|
1690 |
+
{
|
1691 |
+
"epoch": 1.65,
|
1692 |
+
"learning_rate": 0.0,
|
1693 |
+
"loss": 1.7922,
|
1694 |
+
"step": 4980
|
1695 |
+
},
|
1696 |
+
{
|
1697 |
+
"epoch": 1.66,
|
1698 |
+
"learning_rate": 0.0,
|
1699 |
+
"loss": 1.8409,
|
1700 |
+
"step": 5000
|
1701 |
+
},
|
1702 |
+
{
|
1703 |
+
"epoch": 1.66,
|
1704 |
+
"eval_loss": 1.970503568649292,
|
1705 |
+
"eval_runtime": 11953.6568,
|
1706 |
+
"eval_samples_per_second": 1.118,
|
1707 |
+
"eval_steps_per_second": 0.14,
|
1708 |
+
"step": 5000
|
1709 |
+
},
|
1710 |
+
{
|
1711 |
+
"epoch": 1.67,
|
1712 |
+
"learning_rate": 0.0,
|
1713 |
+
"loss": 1.867,
|
1714 |
+
"step": 5020
|
1715 |
+
},
|
1716 |
+
{
|
1717 |
+
"epoch": 1.67,
|
1718 |
+
"learning_rate": 0.0,
|
1719 |
+
"loss": 1.8902,
|
1720 |
+
"step": 5040
|
1721 |
+
},
|
1722 |
+
{
|
1723 |
+
"epoch": 1.68,
|
1724 |
+
"learning_rate": 0.0,
|
1725 |
+
"loss": 1.8835,
|
1726 |
+
"step": 5060
|
1727 |
+
},
|
1728 |
+
{
|
1729 |
+
"epoch": 1.68,
|
1730 |
+
"learning_rate": 0.0,
|
1731 |
+
"loss": 1.741,
|
1732 |
+
"step": 5080
|
1733 |
+
},
|
1734 |
+
{
|
1735 |
+
"epoch": 1.69,
|
1736 |
+
"learning_rate": 0.0,
|
1737 |
+
"loss": 1.7543,
|
1738 |
+
"step": 5100
|
1739 |
+
},
|
1740 |
+
{
|
1741 |
+
"epoch": 1.7,
|
1742 |
+
"learning_rate": 0.0,
|
1743 |
+
"loss": 1.7913,
|
1744 |
+
"step": 5120
|
1745 |
+
},
|
1746 |
+
{
|
1747 |
+
"epoch": 1.7,
|
1748 |
+
"learning_rate": 0.0,
|
1749 |
+
"loss": 1.829,
|
1750 |
+
"step": 5140
|
1751 |
+
},
|
1752 |
+
{
|
1753 |
+
"epoch": 1.71,
|
1754 |
+
"learning_rate": 0.0,
|
1755 |
+
"loss": 1.7915,
|
1756 |
+
"step": 5160
|
1757 |
+
},
|
1758 |
+
{
|
1759 |
+
"epoch": 1.72,
|
1760 |
+
"learning_rate": 0.0,
|
1761 |
+
"loss": 1.767,
|
1762 |
+
"step": 5180
|
1763 |
+
},
|
1764 |
+
{
|
1765 |
+
"epoch": 1.72,
|
1766 |
+
"learning_rate": 0.0,
|
1767 |
+
"loss": 1.7809,
|
1768 |
+
"step": 5200
|
1769 |
+
},
|
1770 |
+
{
|
1771 |
+
"epoch": 1.72,
|
1772 |
+
"eval_loss": 1.970503568649292,
|
1773 |
+
"eval_runtime": 11951.295,
|
1774 |
+
"eval_samples_per_second": 1.119,
|
1775 |
+
"eval_steps_per_second": 0.14,
|
1776 |
+
"step": 5200
|
1777 |
+
},
|
1778 |
+
{
|
1779 |
+
"epoch": 1.73,
|
1780 |
+
"learning_rate": 0.0,
|
1781 |
+
"loss": 1.8715,
|
1782 |
+
"step": 5220
|
1783 |
+
},
|
1784 |
+
{
|
1785 |
+
"epoch": 1.74,
|
1786 |
+
"learning_rate": 0.0,
|
1787 |
+
"loss": 1.8517,
|
1788 |
+
"step": 5240
|
1789 |
+
},
|
1790 |
+
{
|
1791 |
+
"epoch": 1.74,
|
1792 |
+
"learning_rate": 0.0,
|
1793 |
+
"loss": 1.8255,
|
1794 |
+
"step": 5260
|
1795 |
+
},
|
1796 |
+
{
|
1797 |
+
"epoch": 1.75,
|
1798 |
+
"learning_rate": 0.0,
|
1799 |
+
"loss": 1.7973,
|
1800 |
+
"step": 5280
|
1801 |
+
},
|
1802 |
+
{
|
1803 |
+
"epoch": 1.76,
|
1804 |
+
"learning_rate": 0.0,
|
1805 |
+
"loss": 1.7912,
|
1806 |
+
"step": 5300
|
1807 |
+
},
|
1808 |
+
{
|
1809 |
+
"epoch": 1.76,
|
1810 |
+
"learning_rate": 0.0,
|
1811 |
+
"loss": 1.8331,
|
1812 |
+
"step": 5320
|
1813 |
+
},
|
1814 |
+
{
|
1815 |
+
"epoch": 1.77,
|
1816 |
+
"learning_rate": 0.0,
|
1817 |
+
"loss": 1.835,
|
1818 |
+
"step": 5340
|
1819 |
+
},
|
1820 |
+
{
|
1821 |
+
"epoch": 1.78,
|
1822 |
+
"learning_rate": 0.0,
|
1823 |
+
"loss": 1.8051,
|
1824 |
+
"step": 5360
|
1825 |
+
},
|
1826 |
+
{
|
1827 |
+
"epoch": 1.78,
|
1828 |
+
"learning_rate": 0.0,
|
1829 |
+
"loss": 1.8007,
|
1830 |
+
"step": 5380
|
1831 |
+
},
|
1832 |
+
{
|
1833 |
+
"epoch": 1.79,
|
1834 |
+
"learning_rate": 0.0,
|
1835 |
+
"loss": 1.8069,
|
1836 |
+
"step": 5400
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 1.79,
|
1840 |
+
"eval_loss": 1.970503568649292,
|
1841 |
+
"eval_runtime": 11948.5086,
|
1842 |
+
"eval_samples_per_second": 1.119,
|
1843 |
+
"eval_steps_per_second": 0.14,
|
1844 |
+
"step": 5400
|
1845 |
+
},
|
1846 |
+
{
|
1847 |
+
"epoch": 1.8,
|
1848 |
+
"learning_rate": 0.0,
|
1849 |
+
"loss": 1.784,
|
1850 |
+
"step": 5420
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 1.8,
|
1854 |
+
"learning_rate": 0.0,
|
1855 |
+
"loss": 1.7557,
|
1856 |
+
"step": 5440
|
1857 |
+
},
|
1858 |
+
{
|
1859 |
+
"epoch": 1.81,
|
1860 |
+
"learning_rate": 0.0,
|
1861 |
+
"loss": 1.8011,
|
1862 |
+
"step": 5460
|
1863 |
+
},
|
1864 |
+
{
|
1865 |
+
"epoch": 1.82,
|
1866 |
+
"learning_rate": 0.0,
|
1867 |
+
"loss": 1.8392,
|
1868 |
+
"step": 5480
|
1869 |
+
},
|
1870 |
+
{
|
1871 |
+
"epoch": 1.82,
|
1872 |
+
"learning_rate": 0.0,
|
1873 |
+
"loss": 1.7507,
|
1874 |
+
"step": 5500
|
1875 |
+
},
|
1876 |
+
{
|
1877 |
+
"epoch": 1.83,
|
1878 |
+
"learning_rate": 0.0,
|
1879 |
+
"loss": 1.8043,
|
1880 |
+
"step": 5520
|
1881 |
+
},
|
1882 |
+
{
|
1883 |
+
"epoch": 1.84,
|
1884 |
+
"learning_rate": 0.0,
|
1885 |
+
"loss": 1.7804,
|
1886 |
+
"step": 5540
|
1887 |
+
},
|
1888 |
+
{
|
1889 |
+
"epoch": 1.84,
|
1890 |
+
"learning_rate": 0.0,
|
1891 |
+
"loss": 1.811,
|
1892 |
+
"step": 5560
|
1893 |
+
},
|
1894 |
+
{
|
1895 |
+
"epoch": 1.85,
|
1896 |
+
"learning_rate": 0.0,
|
1897 |
+
"loss": 1.806,
|
1898 |
+
"step": 5580
|
1899 |
+
},
|
1900 |
+
{
|
1901 |
+
"epoch": 1.86,
|
1902 |
+
"learning_rate": 0.0,
|
1903 |
+
"loss": 1.8103,
|
1904 |
+
"step": 5600
|
1905 |
+
},
|
1906 |
+
{
|
1907 |
+
"epoch": 1.86,
|
1908 |
+
"eval_loss": 1.970503568649292,
|
1909 |
+
"eval_runtime": 11953.5582,
|
1910 |
+
"eval_samples_per_second": 1.118,
|
1911 |
+
"eval_steps_per_second": 0.14,
|
1912 |
+
"step": 5600
|
1913 |
+
},
|
1914 |
+
{
|
1915 |
+
"epoch": 1.86,
|
1916 |
+
"learning_rate": 0.0,
|
1917 |
+
"loss": 1.7947,
|
1918 |
+
"step": 5620
|
1919 |
+
},
|
1920 |
+
{
|
1921 |
+
"epoch": 1.87,
|
1922 |
+
"learning_rate": 0.0,
|
1923 |
+
"loss": 1.8062,
|
1924 |
+
"step": 5640
|
1925 |
+
},
|
1926 |
+
{
|
1927 |
+
"epoch": 1.88,
|
1928 |
+
"learning_rate": 0.0,
|
1929 |
+
"loss": 1.7997,
|
1930 |
+
"step": 5660
|
1931 |
+
},
|
1932 |
+
{
|
1933 |
+
"epoch": 1.88,
|
1934 |
+
"learning_rate": 0.0,
|
1935 |
+
"loss": 1.843,
|
1936 |
+
"step": 5680
|
1937 |
+
},
|
1938 |
+
{
|
1939 |
+
"epoch": 1.89,
|
1940 |
+
"learning_rate": 0.0,
|
1941 |
+
"loss": 1.848,
|
1942 |
+
"step": 5700
|
1943 |
+
},
|
1944 |
+
{
|
1945 |
+
"epoch": 1.9,
|
1946 |
+
"learning_rate": 0.0,
|
1947 |
+
"loss": 1.8393,
|
1948 |
+
"step": 5720
|
1949 |
+
},
|
1950 |
+
{
|
1951 |
+
"epoch": 1.9,
|
1952 |
+
"learning_rate": 0.0,
|
1953 |
+
"loss": 1.8187,
|
1954 |
+
"step": 5740
|
1955 |
+
},
|
1956 |
+
{
|
1957 |
+
"epoch": 1.91,
|
1958 |
+
"learning_rate": 0.0,
|
1959 |
+
"loss": 1.802,
|
1960 |
+
"step": 5760
|
1961 |
+
},
|
1962 |
+
{
|
1963 |
+
"epoch": 1.92,
|
1964 |
+
"learning_rate": 0.0,
|
1965 |
+
"loss": 1.7646,
|
1966 |
+
"step": 5780
|
1967 |
+
},
|
1968 |
+
{
|
1969 |
+
"epoch": 1.92,
|
1970 |
+
"learning_rate": 0.0,
|
1971 |
+
"loss": 1.7993,
|
1972 |
+
"step": 5800
|
1973 |
+
},
|
1974 |
+
{
|
1975 |
+
"epoch": 1.92,
|
1976 |
+
"eval_loss": 1.970503568649292,
|
1977 |
+
"eval_runtime": 11953.0956,
|
1978 |
+
"eval_samples_per_second": 1.118,
|
1979 |
+
"eval_steps_per_second": 0.14,
|
1980 |
+
"step": 5800
|
1981 |
+
},
|
1982 |
+
{
|
1983 |
+
"epoch": 1.93,
|
1984 |
+
"learning_rate": 0.0,
|
1985 |
+
"loss": 1.7962,
|
1986 |
+
"step": 5820
|
1987 |
+
},
|
1988 |
+
{
|
1989 |
+
"epoch": 1.94,
|
1990 |
+
"learning_rate": 0.0,
|
1991 |
+
"loss": 1.849,
|
1992 |
+
"step": 5840
|
1993 |
+
},
|
1994 |
+
{
|
1995 |
+
"epoch": 1.94,
|
1996 |
+
"learning_rate": 0.0,
|
1997 |
+
"loss": 1.8254,
|
1998 |
+
"step": 5860
|
1999 |
+
},
|
2000 |
+
{
|
2001 |
+
"epoch": 1.95,
|
2002 |
+
"learning_rate": 0.0,
|
2003 |
+
"loss": 1.8583,
|
2004 |
+
"step": 5880
|
2005 |
+
},
|
2006 |
+
{
|
2007 |
+
"epoch": 1.96,
|
2008 |
+
"learning_rate": 0.0,
|
2009 |
+
"loss": 1.8398,
|
2010 |
+
"step": 5900
|
2011 |
+
},
|
2012 |
+
{
|
2013 |
+
"epoch": 1.96,
|
2014 |
+
"learning_rate": 0.0,
|
2015 |
+
"loss": 1.7306,
|
2016 |
+
"step": 5920
|
2017 |
+
},
|
2018 |
+
{
|
2019 |
+
"epoch": 1.97,
|
2020 |
+
"learning_rate": 0.0,
|
2021 |
+
"loss": 1.81,
|
2022 |
+
"step": 5940
|
2023 |
+
},
|
2024 |
+
{
|
2025 |
+
"epoch": 1.98,
|
2026 |
+
"learning_rate": 0.0,
|
2027 |
+
"loss": 1.7762,
|
2028 |
+
"step": 5960
|
2029 |
+
},
|
2030 |
+
{
|
2031 |
+
"epoch": 1.98,
|
2032 |
+
"learning_rate": 0.0,
|
2033 |
+
"loss": 1.8549,
|
2034 |
+
"step": 5980
|
2035 |
+
},
|
2036 |
+
{
|
2037 |
+
"epoch": 1.99,
|
2038 |
+
"learning_rate": 0.0,
|
2039 |
+
"loss": 1.84,
|
2040 |
+
"step": 6000
|
2041 |
+
},
|
2042 |
+
{
|
2043 |
+
"epoch": 1.99,
|
2044 |
+
"eval_loss": 1.970503568649292,
|
2045 |
+
"eval_runtime": 11950.3766,
|
2046 |
+
"eval_samples_per_second": 1.119,
|
2047 |
+
"eval_steps_per_second": 0.14,
|
2048 |
+
"step": 6000
|
2049 |
+
},
|
2050 |
+
{
|
2051 |
+
"epoch": 2.0,
|
2052 |
+
"learning_rate": 0.0,
|
2053 |
+
"loss": 1.7897,
|
2054 |
+
"step": 6020
|
2055 |
+
},
|
2056 |
+
{
|
2057 |
+
"epoch": 2.0,
|
2058 |
+
"learning_rate": 0.0,
|
2059 |
+
"loss": 1.7844,
|
2060 |
+
"step": 6040
|
2061 |
+
},
|
2062 |
+
{
|
2063 |
+
"epoch": 2.01,
|
2064 |
+
"learning_rate": 0.0,
|
2065 |
+
"loss": 1.8453,
|
2066 |
+
"step": 6060
|
2067 |
+
},
|
2068 |
+
{
|
2069 |
+
"epoch": 2.02,
|
2070 |
+
"learning_rate": 0.0,
|
2071 |
+
"loss": 1.8268,
|
2072 |
+
"step": 6080
|
2073 |
+
},
|
2074 |
+
{
|
2075 |
+
"epoch": 2.02,
|
2076 |
+
"learning_rate": 0.0,
|
2077 |
+
"loss": 1.7932,
|
2078 |
+
"step": 6100
|
2079 |
+
},
|
2080 |
+
{
|
2081 |
+
"epoch": 2.03,
|
2082 |
+
"learning_rate": 0.0,
|
2083 |
+
"loss": 1.8005,
|
2084 |
+
"step": 6120
|
2085 |
+
},
|
2086 |
+
{
|
2087 |
+
"epoch": 2.04,
|
2088 |
+
"learning_rate": 0.0,
|
2089 |
+
"loss": 1.773,
|
2090 |
+
"step": 6140
|
2091 |
+
},
|
2092 |
+
{
|
2093 |
+
"epoch": 2.04,
|
2094 |
+
"learning_rate": 0.0,
|
2095 |
+
"loss": 1.8029,
|
2096 |
+
"step": 6160
|
2097 |
+
},
|
2098 |
+
{
|
2099 |
+
"epoch": 2.05,
|
2100 |
+
"learning_rate": 0.0,
|
2101 |
+
"loss": 1.8283,
|
2102 |
+
"step": 6180
|
2103 |
+
},
|
2104 |
+
{
|
2105 |
+
"epoch": 2.06,
|
2106 |
+
"learning_rate": 0.0,
|
2107 |
+
"loss": 1.8167,
|
2108 |
+
"step": 6200
|
2109 |
+
},
|
2110 |
+
{
|
2111 |
+
"epoch": 2.06,
|
2112 |
+
"eval_loss": 1.970503568649292,
|
2113 |
+
"eval_runtime": 11950.9199,
|
2114 |
+
"eval_samples_per_second": 1.119,
|
2115 |
+
"eval_steps_per_second": 0.14,
|
2116 |
+
"step": 6200
|
2117 |
+
},
|
2118 |
+
{
|
2119 |
+
"epoch": 2.06,
|
2120 |
+
"learning_rate": 0.0,
|
2121 |
+
"loss": 1.8465,
|
2122 |
+
"step": 6220
|
2123 |
+
},
|
2124 |
+
{
|
2125 |
+
"epoch": 2.07,
|
2126 |
+
"learning_rate": 0.0,
|
2127 |
+
"loss": 1.8747,
|
2128 |
+
"step": 6240
|
2129 |
+
},
|
2130 |
+
{
|
2131 |
+
"epoch": 2.08,
|
2132 |
+
"learning_rate": 0.0,
|
2133 |
+
"loss": 1.8031,
|
2134 |
+
"step": 6260
|
2135 |
+
},
|
2136 |
+
{
|
2137 |
+
"epoch": 2.08,
|
2138 |
+
"learning_rate": 0.0,
|
2139 |
+
"loss": 1.8366,
|
2140 |
+
"step": 6280
|
2141 |
+
},
|
2142 |
+
{
|
2143 |
+
"epoch": 2.09,
|
2144 |
+
"learning_rate": 0.0,
|
2145 |
+
"loss": 1.7998,
|
2146 |
+
"step": 6300
|
2147 |
+
},
|
2148 |
+
{
|
2149 |
+
"epoch": 2.1,
|
2150 |
+
"learning_rate": 0.0,
|
2151 |
+
"loss": 1.8143,
|
2152 |
+
"step": 6320
|
2153 |
+
},
|
2154 |
+
{
|
2155 |
+
"epoch": 2.1,
|
2156 |
+
"learning_rate": 0.0,
|
2157 |
+
"loss": 1.8586,
|
2158 |
+
"step": 6340
|
2159 |
+
},
|
2160 |
+
{
|
2161 |
+
"epoch": 2.11,
|
2162 |
+
"learning_rate": 0.0,
|
2163 |
+
"loss": 1.836,
|
2164 |
+
"step": 6360
|
2165 |
+
},
|
2166 |
+
{
|
2167 |
+
"epoch": 2.12,
|
2168 |
+
"learning_rate": 0.0,
|
2169 |
+
"loss": 1.863,
|
2170 |
+
"step": 6380
|
2171 |
+
},
|
2172 |
+
{
|
2173 |
+
"epoch": 2.12,
|
2174 |
+
"learning_rate": 0.0,
|
2175 |
+
"loss": 1.7667,
|
2176 |
+
"step": 6400
|
2177 |
+
},
|
2178 |
+
{
|
2179 |
+
"epoch": 2.12,
|
2180 |
+
"eval_loss": 1.970503568649292,
|
2181 |
+
"eval_runtime": 11949.7143,
|
2182 |
+
"eval_samples_per_second": 1.119,
|
2183 |
+
"eval_steps_per_second": 0.14,
|
2184 |
+
"step": 6400
|
2185 |
+
},
|
2186 |
+
{
|
2187 |
+
"epoch": 2.13,
|
2188 |
+
"learning_rate": 0.0,
|
2189 |
+
"loss": 1.8187,
|
2190 |
+
"step": 6420
|
2191 |
+
},
|
2192 |
+
{
|
2193 |
+
"epoch": 2.14,
|
2194 |
+
"learning_rate": 0.0,
|
2195 |
+
"loss": 1.7883,
|
2196 |
+
"step": 6440
|
2197 |
+
},
|
2198 |
+
{
|
2199 |
+
"epoch": 2.14,
|
2200 |
+
"learning_rate": 0.0,
|
2201 |
+
"loss": 1.9299,
|
2202 |
+
"step": 6460
|
2203 |
+
},
|
2204 |
+
{
|
2205 |
+
"epoch": 2.15,
|
2206 |
+
"learning_rate": 0.0,
|
2207 |
+
"loss": 1.8286,
|
2208 |
+
"step": 6480
|
2209 |
+
},
|
2210 |
+
{
|
2211 |
+
"epoch": 2.16,
|
2212 |
+
"learning_rate": 0.0,
|
2213 |
+
"loss": 1.8181,
|
2214 |
+
"step": 6500
|
2215 |
+
},
|
2216 |
+
{
|
2217 |
+
"epoch": 2.16,
|
2218 |
+
"learning_rate": 0.0,
|
2219 |
+
"loss": 1.7856,
|
2220 |
+
"step": 6520
|
2221 |
+
},
|
2222 |
+
{
|
2223 |
+
"epoch": 2.17,
|
2224 |
+
"learning_rate": 0.0,
|
2225 |
+
"loss": 1.7192,
|
2226 |
+
"step": 6540
|
2227 |
+
},
|
2228 |
+
{
|
2229 |
+
"epoch": 2.18,
|
2230 |
+
"learning_rate": 0.0,
|
2231 |
+
"loss": 1.7715,
|
2232 |
+
"step": 6560
|
2233 |
+
},
|
2234 |
+
{
|
2235 |
+
"epoch": 2.18,
|
2236 |
+
"learning_rate": 0.0,
|
2237 |
+
"loss": 1.8359,
|
2238 |
+
"step": 6580
|
2239 |
+
},
|
2240 |
+
{
|
2241 |
+
"epoch": 2.19,
|
2242 |
+
"learning_rate": 0.0,
|
2243 |
+
"loss": 1.7989,
|
2244 |
+
"step": 6600
|
2245 |
+
},
|
2246 |
+
{
|
2247 |
+
"epoch": 2.19,
|
2248 |
+
"eval_loss": 1.970503568649292,
|
2249 |
+
"eval_runtime": 11949.8024,
|
2250 |
+
"eval_samples_per_second": 1.119,
|
2251 |
+
"eval_steps_per_second": 0.14,
|
2252 |
+
"step": 6600
|
2253 |
+
},
|
2254 |
+
{
|
2255 |
+
"epoch": 2.2,
|
2256 |
+
"learning_rate": 0.0,
|
2257 |
+
"loss": 1.7869,
|
2258 |
+
"step": 6620
|
2259 |
+
},
|
2260 |
+
{
|
2261 |
+
"epoch": 2.2,
|
2262 |
+
"learning_rate": 0.0,
|
2263 |
+
"loss": 1.7855,
|
2264 |
+
"step": 6640
|
2265 |
+
},
|
2266 |
+
{
|
2267 |
+
"epoch": 2.21,
|
2268 |
+
"learning_rate": 0.0,
|
2269 |
+
"loss": 1.8312,
|
2270 |
+
"step": 6660
|
2271 |
+
},
|
2272 |
+
{
|
2273 |
+
"epoch": 2.22,
|
2274 |
+
"learning_rate": 0.0,
|
2275 |
+
"loss": 1.7717,
|
2276 |
+
"step": 6680
|
2277 |
+
},
|
2278 |
+
{
|
2279 |
+
"epoch": 2.22,
|
2280 |
+
"learning_rate": 0.0,
|
2281 |
+
"loss": 1.7621,
|
2282 |
+
"step": 6700
|
2283 |
+
},
|
2284 |
+
{
|
2285 |
+
"epoch": 2.23,
|
2286 |
+
"learning_rate": 0.0,
|
2287 |
+
"loss": 1.8031,
|
2288 |
+
"step": 6720
|
2289 |
+
},
|
2290 |
+
{
|
2291 |
+
"epoch": 2.24,
|
2292 |
+
"learning_rate": 0.0,
|
2293 |
+
"loss": 1.8061,
|
2294 |
+
"step": 6740
|
2295 |
+
},
|
2296 |
+
{
|
2297 |
+
"epoch": 2.24,
|
2298 |
+
"learning_rate": 0.0,
|
2299 |
+
"loss": 1.8063,
|
2300 |
+
"step": 6760
|
2301 |
+
},
|
2302 |
+
{
|
2303 |
+
"epoch": 2.25,
|
2304 |
+
"learning_rate": 0.0,
|
2305 |
+
"loss": 1.7469,
|
2306 |
+
"step": 6780
|
2307 |
+
},
|
2308 |
+
{
|
2309 |
+
"epoch": 2.26,
|
2310 |
+
"learning_rate": 0.0,
|
2311 |
+
"loss": 1.7543,
|
2312 |
+
"step": 6800
|
2313 |
+
},
|
2314 |
+
{
|
2315 |
+
"epoch": 2.26,
|
2316 |
+
"eval_loss": 1.970503568649292,
|
2317 |
+
"eval_runtime": 11954.2042,
|
2318 |
+
"eval_samples_per_second": 1.118,
|
2319 |
+
"eval_steps_per_second": 0.14,
|
2320 |
+
"step": 6800
|
2321 |
+
},
|
2322 |
+
{
|
2323 |
+
"epoch": 2.26,
|
2324 |
+
"learning_rate": 0.0,
|
2325 |
+
"loss": 1.8659,
|
2326 |
+
"step": 6820
|
2327 |
+
},
|
2328 |
+
{
|
2329 |
+
"epoch": 2.27,
|
2330 |
+
"learning_rate": 0.0,
|
2331 |
+
"loss": 1.7935,
|
2332 |
+
"step": 6840
|
2333 |
+
},
|
2334 |
+
{
|
2335 |
+
"epoch": 2.28,
|
2336 |
+
"learning_rate": 0.0,
|
2337 |
+
"loss": 1.8247,
|
2338 |
+
"step": 6860
|
2339 |
+
},
|
2340 |
+
{
|
2341 |
+
"epoch": 2.28,
|
2342 |
+
"learning_rate": 0.0,
|
2343 |
+
"loss": 1.8339,
|
2344 |
+
"step": 6880
|
2345 |
+
},
|
2346 |
+
{
|
2347 |
+
"epoch": 2.29,
|
2348 |
+
"learning_rate": 0.0,
|
2349 |
+
"loss": 1.8419,
|
2350 |
+
"step": 6900
|
2351 |
+
},
|
2352 |
+
{
|
2353 |
+
"epoch": 2.3,
|
2354 |
+
"learning_rate": 0.0,
|
2355 |
+
"loss": 1.7978,
|
2356 |
+
"step": 6920
|
2357 |
+
},
|
2358 |
+
{
|
2359 |
+
"epoch": 2.3,
|
2360 |
+
"learning_rate": 0.0,
|
2361 |
+
"loss": 1.8048,
|
2362 |
+
"step": 6940
|
2363 |
+
},
|
2364 |
+
{
|
2365 |
+
"epoch": 2.31,
|
2366 |
+
"learning_rate": 0.0,
|
2367 |
+
"loss": 1.8523,
|
2368 |
+
"step": 6960
|
2369 |
+
},
|
2370 |
+
{
|
2371 |
+
"epoch": 2.32,
|
2372 |
+
"learning_rate": 0.0,
|
2373 |
+
"loss": 1.7875,
|
2374 |
+
"step": 6980
|
2375 |
+
},
|
2376 |
+
{
|
2377 |
+
"epoch": 2.32,
|
2378 |
+
"learning_rate": 0.0,
|
2379 |
+
"loss": 1.8262,
|
2380 |
+
"step": 7000
|
2381 |
+
},
|
2382 |
+
{
|
2383 |
+
"epoch": 2.32,
|
2384 |
+
"eval_loss": 1.970503568649292,
|
2385 |
+
"eval_runtime": 11967.8262,
|
2386 |
+
"eval_samples_per_second": 1.117,
|
2387 |
+
"eval_steps_per_second": 0.14,
|
2388 |
+
"step": 7000
|
2389 |
+
},
|
2390 |
+
{
|
2391 |
+
"epoch": 2.33,
|
2392 |
+
"learning_rate": 0.0,
|
2393 |
+
"loss": 1.8661,
|
2394 |
+
"step": 7020
|
2395 |
+
},
|
2396 |
+
{
|
2397 |
+
"epoch": 2.33,
|
2398 |
+
"learning_rate": 0.0,
|
2399 |
+
"loss": 1.8136,
|
2400 |
+
"step": 7040
|
2401 |
+
},
|
2402 |
+
{
|
2403 |
+
"epoch": 2.34,
|
2404 |
+
"learning_rate": 0.0,
|
2405 |
+
"loss": 1.8066,
|
2406 |
+
"step": 7060
|
2407 |
+
},
|
2408 |
+
{
|
2409 |
+
"epoch": 2.35,
|
2410 |
+
"learning_rate": 0.0,
|
2411 |
+
"loss": 1.8355,
|
2412 |
+
"step": 7080
|
2413 |
+
},
|
2414 |
+
{
|
2415 |
+
"epoch": 2.35,
|
2416 |
+
"learning_rate": 0.0,
|
2417 |
+
"loss": 1.7598,
|
2418 |
+
"step": 7100
|
2419 |
+
},
|
2420 |
+
{
|
2421 |
+
"epoch": 2.36,
|
2422 |
+
"learning_rate": 0.0,
|
2423 |
+
"loss": 1.8384,
|
2424 |
+
"step": 7120
|
2425 |
+
},
|
2426 |
+
{
|
2427 |
+
"epoch": 2.37,
|
2428 |
+
"learning_rate": 0.0,
|
2429 |
+
"loss": 1.7768,
|
2430 |
+
"step": 7140
|
2431 |
+
},
|
2432 |
+
{
|
2433 |
+
"epoch": 2.37,
|
2434 |
+
"learning_rate": 0.0,
|
2435 |
+
"loss": 1.8371,
|
2436 |
+
"step": 7160
|
2437 |
+
},
|
2438 |
+
{
|
2439 |
+
"epoch": 2.38,
|
2440 |
+
"learning_rate": 0.0,
|
2441 |
+
"loss": 1.7989,
|
2442 |
+
"step": 7180
|
2443 |
+
},
|
2444 |
+
{
|
2445 |
+
"epoch": 2.39,
|
2446 |
+
"learning_rate": 0.0,
|
2447 |
+
"loss": 1.8204,
|
2448 |
+
"step": 7200
|
2449 |
+
},
|
2450 |
+
{
|
2451 |
+
"epoch": 2.39,
|
2452 |
+
"eval_loss": 1.970503568649292,
|
2453 |
+
"eval_runtime": 11959.9088,
|
2454 |
+
"eval_samples_per_second": 1.118,
|
2455 |
+
"eval_steps_per_second": 0.14,
|
2456 |
+
"step": 7200
|
2457 |
+
},
|
2458 |
+
{
|
2459 |
+
"epoch": 2.39,
|
2460 |
+
"learning_rate": 0.0,
|
2461 |
+
"loss": 1.8632,
|
2462 |
+
"step": 7220
|
2463 |
+
},
|
2464 |
+
{
|
2465 |
+
"epoch": 2.4,
|
2466 |
+
"learning_rate": 0.0,
|
2467 |
+
"loss": 1.838,
|
2468 |
+
"step": 7240
|
2469 |
+
},
|
2470 |
+
{
|
2471 |
+
"epoch": 2.41,
|
2472 |
+
"learning_rate": 0.0,
|
2473 |
+
"loss": 1.8492,
|
2474 |
+
"step": 7260
|
2475 |
+
},
|
2476 |
+
{
|
2477 |
+
"epoch": 2.41,
|
2478 |
+
"learning_rate": 0.0,
|
2479 |
+
"loss": 1.8213,
|
2480 |
+
"step": 7280
|
2481 |
+
},
|
2482 |
+
{
|
2483 |
+
"epoch": 2.42,
|
2484 |
+
"learning_rate": 0.0,
|
2485 |
+
"loss": 1.7367,
|
2486 |
+
"step": 7300
|
2487 |
+
},
|
2488 |
+
{
|
2489 |
+
"epoch": 2.43,
|
2490 |
+
"learning_rate": 0.0,
|
2491 |
+
"loss": 1.9046,
|
2492 |
+
"step": 7320
|
2493 |
+
},
|
2494 |
+
{
|
2495 |
+
"epoch": 2.43,
|
2496 |
+
"learning_rate": 0.0,
|
2497 |
+
"loss": 1.7799,
|
2498 |
+
"step": 7340
|
2499 |
+
},
|
2500 |
+
{
|
2501 |
+
"epoch": 2.44,
|
2502 |
+
"learning_rate": 0.0,
|
2503 |
+
"loss": 1.793,
|
2504 |
+
"step": 7360
|
2505 |
+
},
|
2506 |
+
{
|
2507 |
+
"epoch": 2.45,
|
2508 |
+
"learning_rate": 0.0,
|
2509 |
+
"loss": 1.7864,
|
2510 |
+
"step": 7380
|
2511 |
+
},
|
2512 |
+
{
|
2513 |
+
"epoch": 2.45,
|
2514 |
+
"learning_rate": 0.0,
|
2515 |
+
"loss": 1.8071,
|
2516 |
+
"step": 7400
|
2517 |
+
},
|
2518 |
+
{
|
2519 |
+
"epoch": 2.45,
|
2520 |
+
"eval_loss": 1.970503568649292,
|
2521 |
+
"eval_runtime": 11949.9374,
|
2522 |
+
"eval_samples_per_second": 1.119,
|
2523 |
+
"eval_steps_per_second": 0.14,
|
2524 |
+
"step": 7400
|
2525 |
+
},
|
2526 |
+
{
|
2527 |
+
"epoch": 2.46,
|
2528 |
+
"learning_rate": 0.0,
|
2529 |
+
"loss": 1.7708,
|
2530 |
+
"step": 7420
|
2531 |
+
},
|
2532 |
+
{
|
2533 |
+
"epoch": 2.47,
|
2534 |
+
"learning_rate": 0.0,
|
2535 |
+
"loss": 1.8234,
|
2536 |
+
"step": 7440
|
2537 |
+
},
|
2538 |
+
{
|
2539 |
+
"epoch": 2.47,
|
2540 |
+
"learning_rate": 0.0,
|
2541 |
+
"loss": 1.8214,
|
2542 |
+
"step": 7460
|
2543 |
+
},
|
2544 |
+
{
|
2545 |
+
"epoch": 2.48,
|
2546 |
+
"learning_rate": 0.0,
|
2547 |
+
"loss": 1.8468,
|
2548 |
+
"step": 7480
|
2549 |
+
},
|
2550 |
+
{
|
2551 |
+
"epoch": 2.49,
|
2552 |
+
"learning_rate": 0.0,
|
2553 |
+
"loss": 1.784,
|
2554 |
+
"step": 7500
|
2555 |
+
},
|
2556 |
+
{
|
2557 |
+
"epoch": 2.49,
|
2558 |
+
"learning_rate": 0.0,
|
2559 |
+
"loss": 1.8571,
|
2560 |
+
"step": 7520
|
2561 |
+
},
|
2562 |
+
{
|
2563 |
+
"epoch": 2.5,
|
2564 |
+
"learning_rate": 0.0,
|
2565 |
+
"loss": 1.7871,
|
2566 |
+
"step": 7540
|
2567 |
+
},
|
2568 |
+
{
|
2569 |
+
"epoch": 2.51,
|
2570 |
+
"learning_rate": 0.0,
|
2571 |
+
"loss": 1.7819,
|
2572 |
+
"step": 7560
|
2573 |
+
},
|
2574 |
+
{
|
2575 |
+
"epoch": 2.51,
|
2576 |
+
"learning_rate": 0.0,
|
2577 |
+
"loss": 1.8035,
|
2578 |
+
"step": 7580
|
2579 |
+
},
|
2580 |
+
{
|
2581 |
+
"epoch": 2.52,
|
2582 |
+
"learning_rate": 0.0,
|
2583 |
+
"loss": 1.801,
|
2584 |
+
"step": 7600
|
2585 |
+
},
|
2586 |
+
{
|
2587 |
+
"epoch": 2.52,
|
2588 |
+
"eval_loss": 1.970503568649292,
|
2589 |
+
"eval_runtime": 11949.3183,
|
2590 |
+
"eval_samples_per_second": 1.119,
|
2591 |
+
"eval_steps_per_second": 0.14,
|
2592 |
+
"step": 7600
|
2593 |
+
},
|
2594 |
+
{
|
2595 |
+
"epoch": 2.53,
|
2596 |
+
"learning_rate": 0.0,
|
2597 |
+
"loss": 1.8302,
|
2598 |
+
"step": 7620
|
2599 |
+
},
|
2600 |
+
{
|
2601 |
+
"epoch": 2.53,
|
2602 |
+
"learning_rate": 0.0,
|
2603 |
+
"loss": 1.8208,
|
2604 |
+
"step": 7640
|
2605 |
+
},
|
2606 |
+
{
|
2607 |
+
"epoch": 2.54,
|
2608 |
+
"learning_rate": 0.0,
|
2609 |
+
"loss": 1.917,
|
2610 |
+
"step": 7660
|
2611 |
+
},
|
2612 |
+
{
|
2613 |
+
"epoch": 2.55,
|
2614 |
+
"learning_rate": 0.0,
|
2615 |
+
"loss": 1.8184,
|
2616 |
+
"step": 7680
|
2617 |
+
},
|
2618 |
+
{
|
2619 |
+
"epoch": 2.55,
|
2620 |
+
"learning_rate": 0.0,
|
2621 |
+
"loss": 1.7462,
|
2622 |
+
"step": 7700
|
2623 |
+
},
|
2624 |
+
{
|
2625 |
+
"epoch": 2.56,
|
2626 |
+
"learning_rate": 0.0,
|
2627 |
+
"loss": 1.7699,
|
2628 |
+
"step": 7720
|
2629 |
+
},
|
2630 |
+
{
|
2631 |
+
"epoch": 2.57,
|
2632 |
+
"learning_rate": 0.0,
|
2633 |
+
"loss": 1.8377,
|
2634 |
+
"step": 7740
|
2635 |
+
},
|
2636 |
+
{
|
2637 |
+
"epoch": 2.57,
|
2638 |
+
"learning_rate": 0.0,
|
2639 |
+
"loss": 1.7713,
|
2640 |
+
"step": 7760
|
2641 |
+
},
|
2642 |
+
{
|
2643 |
+
"epoch": 2.58,
|
2644 |
+
"learning_rate": 0.0,
|
2645 |
+
"loss": 1.798,
|
2646 |
+
"step": 7780
|
2647 |
+
},
|
2648 |
+
{
|
2649 |
+
"epoch": 2.59,
|
2650 |
+
"learning_rate": 0.0,
|
2651 |
+
"loss": 1.8425,
|
2652 |
+
"step": 7800
|
2653 |
+
},
|
2654 |
+
{
|
2655 |
+
"epoch": 2.59,
|
2656 |
+
"eval_loss": 1.970503568649292,
|
2657 |
+
"eval_runtime": 11947.7265,
|
2658 |
+
"eval_samples_per_second": 1.119,
|
2659 |
+
"eval_steps_per_second": 0.14,
|
2660 |
+
"step": 7800
|
2661 |
+
},
|
2662 |
+
{
|
2663 |
+
"epoch": 2.59,
|
2664 |
+
"learning_rate": 0.0,
|
2665 |
+
"loss": 1.7816,
|
2666 |
+
"step": 7820
|
2667 |
+
},
|
2668 |
+
{
|
2669 |
+
"epoch": 2.6,
|
2670 |
+
"learning_rate": 0.0,
|
2671 |
+
"loss": 1.8193,
|
2672 |
+
"step": 7840
|
2673 |
+
},
|
2674 |
+
{
|
2675 |
+
"epoch": 2.61,
|
2676 |
+
"learning_rate": 0.0,
|
2677 |
+
"loss": 1.8044,
|
2678 |
+
"step": 7860
|
2679 |
+
},
|
2680 |
+
{
|
2681 |
+
"epoch": 2.61,
|
2682 |
+
"learning_rate": 0.0,
|
2683 |
+
"loss": 1.7937,
|
2684 |
+
"step": 7880
|
2685 |
+
},
|
2686 |
+
{
|
2687 |
+
"epoch": 2.62,
|
2688 |
+
"learning_rate": 0.0,
|
2689 |
+
"loss": 1.8224,
|
2690 |
+
"step": 7900
|
2691 |
+
},
|
2692 |
+
{
|
2693 |
+
"epoch": 2.63,
|
2694 |
+
"learning_rate": 0.0,
|
2695 |
+
"loss": 1.7937,
|
2696 |
+
"step": 7920
|
2697 |
+
},
|
2698 |
+
{
|
2699 |
+
"epoch": 2.63,
|
2700 |
+
"learning_rate": 0.0,
|
2701 |
+
"loss": 1.7988,
|
2702 |
+
"step": 7940
|
2703 |
+
},
|
2704 |
+
{
|
2705 |
+
"epoch": 2.64,
|
2706 |
+
"learning_rate": 0.0,
|
2707 |
+
"loss": 1.8622,
|
2708 |
+
"step": 7960
|
2709 |
+
},
|
2710 |
+
{
|
2711 |
+
"epoch": 2.65,
|
2712 |
+
"learning_rate": 0.0,
|
2713 |
+
"loss": 1.8407,
|
2714 |
+
"step": 7980
|
2715 |
+
},
|
2716 |
+
{
|
2717 |
+
"epoch": 2.65,
|
2718 |
+
"learning_rate": 0.0,
|
2719 |
+
"loss": 1.7578,
|
2720 |
+
"step": 8000
|
2721 |
+
},
|
2722 |
+
{
|
2723 |
+
"epoch": 2.65,
|
2724 |
+
"eval_loss": 1.970503568649292,
|
2725 |
+
"eval_runtime": 11948.4274,
|
2726 |
+
"eval_samples_per_second": 1.119,
|
2727 |
+
"eval_steps_per_second": 0.14,
|
2728 |
+
"step": 8000
|
2729 |
+
},
|
2730 |
+
{
|
2731 |
+
"epoch": 2.66,
|
2732 |
+
"learning_rate": 0.0,
|
2733 |
+
"loss": 1.789,
|
2734 |
+
"step": 8020
|
2735 |
+
},
|
2736 |
+
{
|
2737 |
+
"epoch": 2.67,
|
2738 |
+
"learning_rate": 0.0,
|
2739 |
+
"loss": 1.7939,
|
2740 |
+
"step": 8040
|
2741 |
+
},
|
2742 |
+
{
|
2743 |
+
"epoch": 2.67,
|
2744 |
+
"learning_rate": 0.0,
|
2745 |
+
"loss": 1.7694,
|
2746 |
+
"step": 8060
|
2747 |
+
},
|
2748 |
+
{
|
2749 |
+
"epoch": 2.68,
|
2750 |
+
"learning_rate": 0.0,
|
2751 |
+
"loss": 1.7917,
|
2752 |
+
"step": 8080
|
2753 |
+
},
|
2754 |
+
{
|
2755 |
+
"epoch": 2.69,
|
2756 |
+
"learning_rate": 0.0,
|
2757 |
+
"loss": 1.7737,
|
2758 |
+
"step": 8100
|
2759 |
+
},
|
2760 |
+
{
|
2761 |
+
"epoch": 2.69,
|
2762 |
+
"learning_rate": 0.0,
|
2763 |
+
"loss": 1.8049,
|
2764 |
+
"step": 8120
|
2765 |
+
},
|
2766 |
+
{
|
2767 |
+
"epoch": 2.7,
|
2768 |
+
"learning_rate": 0.0,
|
2769 |
+
"loss": 1.8419,
|
2770 |
+
"step": 8140
|
2771 |
+
},
|
2772 |
+
{
|
2773 |
+
"epoch": 2.71,
|
2774 |
+
"learning_rate": 0.0,
|
2775 |
+
"loss": 1.7835,
|
2776 |
+
"step": 8160
|
2777 |
+
},
|
2778 |
+
{
|
2779 |
+
"epoch": 2.71,
|
2780 |
+
"learning_rate": 0.0,
|
2781 |
+
"loss": 1.7898,
|
2782 |
+
"step": 8180
|
2783 |
+
},
|
2784 |
+
{
|
2785 |
+
"epoch": 2.72,
|
2786 |
+
"learning_rate": 0.0,
|
2787 |
+
"loss": 1.7798,
|
2788 |
+
"step": 8200
|
2789 |
+
},
|
2790 |
+
{
|
2791 |
+
"epoch": 2.72,
|
2792 |
+
"eval_loss": 1.970503568649292,
|
2793 |
+
"eval_runtime": 11959.3563,
|
2794 |
+
"eval_samples_per_second": 1.118,
|
2795 |
+
"eval_steps_per_second": 0.14,
|
2796 |
+
"step": 8200
|
2797 |
+
},
|
2798 |
+
{
|
2799 |
+
"epoch": 2.73,
|
2800 |
+
"learning_rate": 0.0,
|
2801 |
+
"loss": 1.7904,
|
2802 |
+
"step": 8220
|
2803 |
+
},
|
2804 |
+
{
|
2805 |
+
"epoch": 2.73,
|
2806 |
+
"learning_rate": 0.0,
|
2807 |
+
"loss": 1.8054,
|
2808 |
+
"step": 8240
|
2809 |
+
},
|
2810 |
+
{
|
2811 |
+
"epoch": 2.74,
|
2812 |
+
"learning_rate": 0.0,
|
2813 |
+
"loss": 1.8316,
|
2814 |
+
"step": 8260
|
2815 |
+
},
|
2816 |
+
{
|
2817 |
+
"epoch": 2.75,
|
2818 |
+
"learning_rate": 0.0,
|
2819 |
+
"loss": 1.8059,
|
2820 |
+
"step": 8280
|
2821 |
+
},
|
2822 |
+
{
|
2823 |
+
"epoch": 2.75,
|
2824 |
+
"learning_rate": 0.0,
|
2825 |
+
"loss": 1.7908,
|
2826 |
+
"step": 8300
|
2827 |
+
},
|
2828 |
+
{
|
2829 |
+
"epoch": 2.76,
|
2830 |
+
"learning_rate": 0.0,
|
2831 |
+
"loss": 1.8092,
|
2832 |
+
"step": 8320
|
2833 |
+
},
|
2834 |
+
{
|
2835 |
+
"epoch": 2.77,
|
2836 |
+
"learning_rate": 0.0,
|
2837 |
+
"loss": 1.8954,
|
2838 |
+
"step": 8340
|
2839 |
+
},
|
2840 |
+
{
|
2841 |
+
"epoch": 2.77,
|
2842 |
+
"learning_rate": 0.0,
|
2843 |
+
"loss": 1.835,
|
2844 |
+
"step": 8360
|
2845 |
+
},
|
2846 |
+
{
|
2847 |
+
"epoch": 2.78,
|
2848 |
+
"learning_rate": 0.0,
|
2849 |
+
"loss": 1.8018,
|
2850 |
+
"step": 8380
|
2851 |
+
},
|
2852 |
+
{
|
2853 |
+
"epoch": 2.79,
|
2854 |
+
"learning_rate": 0.0,
|
2855 |
+
"loss": 1.8474,
|
2856 |
+
"step": 8400
|
2857 |
+
},
|
2858 |
+
{
|
2859 |
+
"epoch": 2.79,
|
2860 |
+
"eval_loss": 1.970503568649292,
|
2861 |
+
"eval_runtime": 11978.111,
|
2862 |
+
"eval_samples_per_second": 1.116,
|
2863 |
+
"eval_steps_per_second": 0.14,
|
2864 |
+
"step": 8400
|
2865 |
+
},
|
2866 |
+
{
|
2867 |
+
"epoch": 2.79,
|
2868 |
+
"learning_rate": 0.0,
|
2869 |
+
"loss": 1.7878,
|
2870 |
+
"step": 8420
|
2871 |
+
},
|
2872 |
+
{
|
2873 |
+
"epoch": 2.8,
|
2874 |
+
"learning_rate": 0.0,
|
2875 |
+
"loss": 1.7629,
|
2876 |
+
"step": 8440
|
2877 |
+
},
|
2878 |
+
{
|
2879 |
+
"epoch": 2.81,
|
2880 |
+
"learning_rate": 0.0,
|
2881 |
+
"loss": 1.8068,
|
2882 |
+
"step": 8460
|
2883 |
+
},
|
2884 |
+
{
|
2885 |
+
"epoch": 2.81,
|
2886 |
+
"learning_rate": 0.0,
|
2887 |
+
"loss": 1.7907,
|
2888 |
+
"step": 8480
|
2889 |
+
},
|
2890 |
+
{
|
2891 |
+
"epoch": 2.82,
|
2892 |
+
"learning_rate": 0.0,
|
2893 |
+
"loss": 1.7598,
|
2894 |
+
"step": 8500
|
2895 |
+
},
|
2896 |
+
{
|
2897 |
+
"epoch": 2.83,
|
2898 |
+
"learning_rate": 0.0,
|
2899 |
+
"loss": 1.7964,
|
2900 |
+
"step": 8520
|
2901 |
+
},
|
2902 |
+
{
|
2903 |
+
"epoch": 2.83,
|
2904 |
+
"learning_rate": 0.0,
|
2905 |
+
"loss": 1.7951,
|
2906 |
+
"step": 8540
|
2907 |
+
},
|
2908 |
+
{
|
2909 |
+
"epoch": 2.84,
|
2910 |
+
"learning_rate": 0.0,
|
2911 |
+
"loss": 1.7962,
|
2912 |
+
"step": 8560
|
2913 |
+
},
|
2914 |
+
{
|
2915 |
+
"epoch": 2.85,
|
2916 |
+
"learning_rate": 0.0,
|
2917 |
+
"loss": 1.7908,
|
2918 |
+
"step": 8580
|
2919 |
+
},
|
2920 |
+
{
|
2921 |
+
"epoch": 2.85,
|
2922 |
+
"learning_rate": 0.0,
|
2923 |
+
"loss": 1.7525,
|
2924 |
+
"step": 8600
|
2925 |
+
},
|
2926 |
+
{
|
2927 |
+
"epoch": 2.85,
|
2928 |
+
"eval_loss": 1.970503568649292,
|
2929 |
+
"eval_runtime": 11977.8941,
|
2930 |
+
"eval_samples_per_second": 1.116,
|
2931 |
+
"eval_steps_per_second": 0.14,
|
2932 |
+
"step": 8600
|
2933 |
+
},
|
2934 |
+
{
|
2935 |
+
"epoch": 2.86,
|
2936 |
+
"learning_rate": 0.0,
|
2937 |
+
"loss": 1.7268,
|
2938 |
+
"step": 8620
|
2939 |
+
},
|
2940 |
+
{
|
2941 |
+
"epoch": 2.87,
|
2942 |
+
"learning_rate": 0.0,
|
2943 |
+
"loss": 1.7716,
|
2944 |
+
"step": 8640
|
2945 |
+
},
|
2946 |
+
{
|
2947 |
+
"epoch": 2.87,
|
2948 |
+
"learning_rate": 0.0,
|
2949 |
+
"loss": 1.8214,
|
2950 |
+
"step": 8660
|
2951 |
+
},
|
2952 |
+
{
|
2953 |
+
"epoch": 2.88,
|
2954 |
+
"learning_rate": 0.0,
|
2955 |
+
"loss": 1.8116,
|
2956 |
+
"step": 8680
|
2957 |
+
},
|
2958 |
+
{
|
2959 |
+
"epoch": 2.89,
|
2960 |
+
"learning_rate": 0.0,
|
2961 |
+
"loss": 1.8204,
|
2962 |
+
"step": 8700
|
2963 |
+
},
|
2964 |
+
{
|
2965 |
+
"epoch": 2.89,
|
2966 |
+
"learning_rate": 0.0,
|
2967 |
+
"loss": 1.7878,
|
2968 |
+
"step": 8720
|
2969 |
+
},
|
2970 |
+
{
|
2971 |
+
"epoch": 2.9,
|
2972 |
+
"learning_rate": 0.0,
|
2973 |
+
"loss": 1.8828,
|
2974 |
+
"step": 8740
|
2975 |
+
},
|
2976 |
+
{
|
2977 |
+
"epoch": 2.91,
|
2978 |
+
"learning_rate": 0.0,
|
2979 |
+
"loss": 1.8015,
|
2980 |
+
"step": 8760
|
2981 |
+
},
|
2982 |
+
{
|
2983 |
+
"epoch": 2.91,
|
2984 |
+
"learning_rate": 0.0,
|
2985 |
+
"loss": 1.7989,
|
2986 |
+
"step": 8780
|
2987 |
+
},
|
2988 |
+
{
|
2989 |
+
"epoch": 2.92,
|
2990 |
+
"learning_rate": 0.0,
|
2991 |
+
"loss": 1.7467,
|
2992 |
+
"step": 8800
|
2993 |
+
},
|
2994 |
+
{
|
2995 |
+
"epoch": 2.92,
|
2996 |
+
"eval_loss": 1.970503568649292,
|
2997 |
+
"eval_runtime": 11960.8065,
|
2998 |
+
"eval_samples_per_second": 1.118,
|
2999 |
+
"eval_steps_per_second": 0.14,
|
3000 |
+
"step": 8800
|
3001 |
+
}
|
3002 |
+
],
|
3003 |
+
"max_steps": 9045,
|
3004 |
+
"num_train_epochs": 3,
|
3005 |
+
"total_flos": 5.4879340491679334e+17,
|
3006 |
+
"trial_name": null,
|
3007 |
+
"trial_params": null
|
3008 |
+
}
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-8800/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f5a78612ea8930d68eba4cb53d62254ccf547582e754aa049d169c3c11dd5fe4
|
3 |
+
size 4027
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-9000/adapter_config.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"base_model_name_or_path": "/domino/edv/afs-mrmc-data-store-rw/innovation/hf/RedPajama-INCITE-7B-Base",
|
3 |
+
"bias": "none",
|
4 |
+
"fan_in_fan_out": false,
|
5 |
+
"inference_mode": true,
|
6 |
+
"init_lora_weights": true,
|
7 |
+
"lora_alpha": 16,
|
8 |
+
"lora_dropout": 0.05,
|
9 |
+
"modules_to_save": null,
|
10 |
+
"peft_type": "LORA",
|
11 |
+
"r": 8,
|
12 |
+
"target_modules": [
|
13 |
+
"query_key_value"
|
14 |
+
],
|
15 |
+
"task_type": "CAUSAL_LM"
|
16 |
+
}
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-9000/adapter_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:36c95157caaa0e8a49516175a38f9e3dfad6634df6c2d7fc47b2e3298cf4d68e
|
3 |
+
size 16800753
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-9000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a3c62c46fbec701f73b9c0d2af49122155f2c1654a0d513a7842e367f0516218
|
3 |
+
size 33592261
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-9000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2d6e839a14f185d11ef5c6540bd37eb7bad7288947ab9f31a50d914db4272e23
|
3 |
+
size 14575
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-9000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:471be078058bbf990ff963153250176cf0c4eef4d5abd363e271020185c5758c
|
3 |
+
size 627
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-9000/trainer_state.json
ADDED
@@ -0,0 +1,3076 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 1.9635850191116333,
|
3 |
+
"best_model_checkpoint": "./results/redpj7B-lora-cnn-dailymail-results_fine_tune_test/checkpoint-600",
|
4 |
+
"epoch": 2.9850746268656714,
|
5 |
+
"global_step": 9000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.01,
|
12 |
+
"learning_rate": 0.00027082228116710874,
|
13 |
+
"loss": 1.8654,
|
14 |
+
"step": 20
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.01,
|
18 |
+
"learning_rate": 0.0002681697612732095,
|
19 |
+
"loss": 1.8124,
|
20 |
+
"step": 40
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.02,
|
24 |
+
"learning_rate": 0.0002655172413793103,
|
25 |
+
"loss": 1.8231,
|
26 |
+
"step": 60
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.03,
|
30 |
+
"learning_rate": 0.0002628647214854111,
|
31 |
+
"loss": 1.7818,
|
32 |
+
"step": 80
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.03,
|
36 |
+
"learning_rate": 0.00026021220159151194,
|
37 |
+
"loss": 1.8446,
|
38 |
+
"step": 100
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.04,
|
42 |
+
"learning_rate": 0.0002575596816976127,
|
43 |
+
"loss": 1.9001,
|
44 |
+
"step": 120
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.05,
|
48 |
+
"learning_rate": 0.0002549071618037135,
|
49 |
+
"loss": 1.8073,
|
50 |
+
"step": 140
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.05,
|
54 |
+
"learning_rate": 0.0002522546419098143,
|
55 |
+
"loss": 1.8506,
|
56 |
+
"step": 160
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.06,
|
60 |
+
"learning_rate": 0.0002496021220159151,
|
61 |
+
"loss": 1.8471,
|
62 |
+
"step": 180
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.07,
|
66 |
+
"learning_rate": 0.0002469496021220159,
|
67 |
+
"loss": 1.8647,
|
68 |
+
"step": 200
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.07,
|
72 |
+
"eval_loss": 1.966022253036499,
|
73 |
+
"eval_runtime": 12002.3952,
|
74 |
+
"eval_samples_per_second": 1.114,
|
75 |
+
"eval_steps_per_second": 0.139,
|
76 |
+
"step": 200
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 0.07,
|
80 |
+
"learning_rate": 0.00024429708222811666,
|
81 |
+
"loss": 1.8578,
|
82 |
+
"step": 220
|
83 |
+
},
|
84 |
+
{
|
85 |
+
"epoch": 0.08,
|
86 |
+
"learning_rate": 0.0002416445623342175,
|
87 |
+
"loss": 1.8329,
|
88 |
+
"step": 240
|
89 |
+
},
|
90 |
+
{
|
91 |
+
"epoch": 0.09,
|
92 |
+
"learning_rate": 0.0002389920424403183,
|
93 |
+
"loss": 1.8119,
|
94 |
+
"step": 260
|
95 |
+
},
|
96 |
+
{
|
97 |
+
"epoch": 0.09,
|
98 |
+
"learning_rate": 0.0002363395225464191,
|
99 |
+
"loss": 1.8884,
|
100 |
+
"step": 280
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.1,
|
104 |
+
"learning_rate": 0.00023368700265251986,
|
105 |
+
"loss": 1.9077,
|
106 |
+
"step": 300
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 0.11,
|
110 |
+
"learning_rate": 0.00023103448275862065,
|
111 |
+
"loss": 1.8092,
|
112 |
+
"step": 320
|
113 |
+
},
|
114 |
+
{
|
115 |
+
"epoch": 0.11,
|
116 |
+
"learning_rate": 0.00022838196286472146,
|
117 |
+
"loss": 1.9237,
|
118 |
+
"step": 340
|
119 |
+
},
|
120 |
+
{
|
121 |
+
"epoch": 0.12,
|
122 |
+
"learning_rate": 0.00022572944297082225,
|
123 |
+
"loss": 1.8701,
|
124 |
+
"step": 360
|
125 |
+
},
|
126 |
+
{
|
127 |
+
"epoch": 0.13,
|
128 |
+
"learning_rate": 0.00022307692307692306,
|
129 |
+
"loss": 1.8933,
|
130 |
+
"step": 380
|
131 |
+
},
|
132 |
+
{
|
133 |
+
"epoch": 0.13,
|
134 |
+
"learning_rate": 0.00022042440318302385,
|
135 |
+
"loss": 1.8453,
|
136 |
+
"step": 400
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.13,
|
140 |
+
"eval_loss": 1.9667036533355713,
|
141 |
+
"eval_runtime": 11991.8036,
|
142 |
+
"eval_samples_per_second": 1.115,
|
143 |
+
"eval_steps_per_second": 0.139,
|
144 |
+
"step": 400
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 0.14,
|
148 |
+
"learning_rate": 0.00021777188328912466,
|
149 |
+
"loss": 1.8272,
|
150 |
+
"step": 420
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 0.15,
|
154 |
+
"learning_rate": 0.00021511936339522545,
|
155 |
+
"loss": 1.8149,
|
156 |
+
"step": 440
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.15,
|
160 |
+
"learning_rate": 0.00021246684350132626,
|
161 |
+
"loss": 1.8565,
|
162 |
+
"step": 460
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 0.16,
|
166 |
+
"learning_rate": 0.00020981432360742705,
|
167 |
+
"loss": 1.8139,
|
168 |
+
"step": 480
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 0.17,
|
172 |
+
"learning_rate": 0.0002071618037135278,
|
173 |
+
"loss": 1.802,
|
174 |
+
"step": 500
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 0.17,
|
178 |
+
"learning_rate": 0.00020450928381962862,
|
179 |
+
"loss": 1.8795,
|
180 |
+
"step": 520
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 0.18,
|
184 |
+
"learning_rate": 0.0002018567639257294,
|
185 |
+
"loss": 1.8621,
|
186 |
+
"step": 540
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 0.19,
|
190 |
+
"learning_rate": 0.00019920424403183022,
|
191 |
+
"loss": 1.8363,
|
192 |
+
"step": 560
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 0.19,
|
196 |
+
"learning_rate": 0.000196551724137931,
|
197 |
+
"loss": 1.8746,
|
198 |
+
"step": 580
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.2,
|
202 |
+
"learning_rate": 0.00019389920424403182,
|
203 |
+
"loss": 1.7763,
|
204 |
+
"step": 600
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 0.2,
|
208 |
+
"eval_loss": 1.9635850191116333,
|
209 |
+
"eval_runtime": 11976.5234,
|
210 |
+
"eval_samples_per_second": 1.116,
|
211 |
+
"eval_steps_per_second": 0.14,
|
212 |
+
"step": 600
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.21,
|
216 |
+
"learning_rate": 0.0001912466843501326,
|
217 |
+
"loss": 1.8423,
|
218 |
+
"step": 620
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 0.21,
|
222 |
+
"learning_rate": 0.00018859416445623343,
|
223 |
+
"loss": 1.8998,
|
224 |
+
"step": 640
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 0.22,
|
228 |
+
"learning_rate": 0.0001859416445623342,
|
229 |
+
"loss": 1.8131,
|
230 |
+
"step": 660
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 0.23,
|
234 |
+
"learning_rate": 0.00018328912466843497,
|
235 |
+
"loss": 1.8656,
|
236 |
+
"step": 680
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.23,
|
240 |
+
"learning_rate": 0.00018063660477453579,
|
241 |
+
"loss": 1.8181,
|
242 |
+
"step": 700
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 0.24,
|
246 |
+
"learning_rate": 0.00017798408488063657,
|
247 |
+
"loss": 1.8425,
|
248 |
+
"step": 720
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.25,
|
252 |
+
"learning_rate": 0.00017533156498673739,
|
253 |
+
"loss": 1.7883,
|
254 |
+
"step": 740
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.25,
|
258 |
+
"learning_rate": 0.00017267904509283817,
|
259 |
+
"loss": 1.8469,
|
260 |
+
"step": 760
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 0.26,
|
264 |
+
"learning_rate": 0.000170026525198939,
|
265 |
+
"loss": 1.8447,
|
266 |
+
"step": 780
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 0.27,
|
270 |
+
"learning_rate": 0.00016737400530503977,
|
271 |
+
"loss": 1.7872,
|
272 |
+
"step": 800
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 0.27,
|
276 |
+
"eval_loss": 1.9704641103744507,
|
277 |
+
"eval_runtime": 11963.7945,
|
278 |
+
"eval_samples_per_second": 1.117,
|
279 |
+
"eval_steps_per_second": 0.14,
|
280 |
+
"step": 800
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 0.27,
|
284 |
+
"learning_rate": 0.0001647214854111406,
|
285 |
+
"loss": 1.8032,
|
286 |
+
"step": 820
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 0.28,
|
290 |
+
"learning_rate": 0.00016206896551724137,
|
291 |
+
"loss": 1.8709,
|
292 |
+
"step": 840
|
293 |
+
},
|
294 |
+
{
|
295 |
+
"epoch": 0.29,
|
296 |
+
"learning_rate": 0.00015941644562334216,
|
297 |
+
"loss": 1.8801,
|
298 |
+
"step": 860
|
299 |
+
},
|
300 |
+
{
|
301 |
+
"epoch": 0.29,
|
302 |
+
"learning_rate": 0.00015676392572944298,
|
303 |
+
"loss": 1.8535,
|
304 |
+
"step": 880
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 0.3,
|
308 |
+
"learning_rate": 0.00015411140583554373,
|
309 |
+
"loss": 1.8631,
|
310 |
+
"step": 900
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.31,
|
314 |
+
"learning_rate": 0.00015145888594164455,
|
315 |
+
"loss": 1.752,
|
316 |
+
"step": 920
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 0.31,
|
320 |
+
"learning_rate": 0.00014880636604774534,
|
321 |
+
"loss": 1.8705,
|
322 |
+
"step": 940
|
323 |
+
},
|
324 |
+
{
|
325 |
+
"epoch": 0.32,
|
326 |
+
"learning_rate": 0.00014615384615384615,
|
327 |
+
"loss": 1.8301,
|
328 |
+
"step": 960
|
329 |
+
},
|
330 |
+
{
|
331 |
+
"epoch": 0.33,
|
332 |
+
"learning_rate": 0.00014350132625994694,
|
333 |
+
"loss": 1.7892,
|
334 |
+
"step": 980
|
335 |
+
},
|
336 |
+
{
|
337 |
+
"epoch": 0.33,
|
338 |
+
"learning_rate": 0.00014084880636604772,
|
339 |
+
"loss": 1.8671,
|
340 |
+
"step": 1000
|
341 |
+
},
|
342 |
+
{
|
343 |
+
"epoch": 0.33,
|
344 |
+
"eval_loss": 1.9677250385284424,
|
345 |
+
"eval_runtime": 11951.074,
|
346 |
+
"eval_samples_per_second": 1.119,
|
347 |
+
"eval_steps_per_second": 0.14,
|
348 |
+
"step": 1000
|
349 |
+
},
|
350 |
+
{
|
351 |
+
"epoch": 0.34,
|
352 |
+
"learning_rate": 0.00013819628647214854,
|
353 |
+
"loss": 1.8168,
|
354 |
+
"step": 1020
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 0.34,
|
358 |
+
"learning_rate": 0.00013554376657824932,
|
359 |
+
"loss": 1.8268,
|
360 |
+
"step": 1040
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 0.35,
|
364 |
+
"learning_rate": 0.00013289124668435014,
|
365 |
+
"loss": 1.8073,
|
366 |
+
"step": 1060
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.36,
|
370 |
+
"learning_rate": 0.0001302387267904509,
|
371 |
+
"loss": 1.9313,
|
372 |
+
"step": 1080
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 0.36,
|
376 |
+
"learning_rate": 0.0001275862068965517,
|
377 |
+
"loss": 1.9097,
|
378 |
+
"step": 1100
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 0.37,
|
382 |
+
"learning_rate": 0.0001249336870026525,
|
383 |
+
"loss": 1.9032,
|
384 |
+
"step": 1120
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 0.38,
|
388 |
+
"learning_rate": 0.0001222811671087533,
|
389 |
+
"loss": 1.8207,
|
390 |
+
"step": 1140
|
391 |
+
},
|
392 |
+
{
|
393 |
+
"epoch": 0.38,
|
394 |
+
"learning_rate": 0.0001196286472148541,
|
395 |
+
"loss": 1.8815,
|
396 |
+
"step": 1160
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 0.39,
|
400 |
+
"learning_rate": 0.0001169761273209549,
|
401 |
+
"loss": 1.8297,
|
402 |
+
"step": 1180
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 0.4,
|
406 |
+
"learning_rate": 0.0001143236074270557,
|
407 |
+
"loss": 1.8899,
|
408 |
+
"step": 1200
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.4,
|
412 |
+
"eval_loss": 1.9722812175750732,
|
413 |
+
"eval_runtime": 11951.1639,
|
414 |
+
"eval_samples_per_second": 1.119,
|
415 |
+
"eval_steps_per_second": 0.14,
|
416 |
+
"step": 1200
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 0.4,
|
420 |
+
"learning_rate": 0.0001116710875331565,
|
421 |
+
"loss": 1.8803,
|
422 |
+
"step": 1220
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.41,
|
426 |
+
"learning_rate": 0.00010901856763925729,
|
427 |
+
"loss": 1.8548,
|
428 |
+
"step": 1240
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 0.42,
|
432 |
+
"learning_rate": 0.00010636604774535807,
|
433 |
+
"loss": 1.8498,
|
434 |
+
"step": 1260
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 0.42,
|
438 |
+
"learning_rate": 0.00010371352785145887,
|
439 |
+
"loss": 1.8057,
|
440 |
+
"step": 1280
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 0.43,
|
444 |
+
"learning_rate": 0.00010106100795755967,
|
445 |
+
"loss": 1.8572,
|
446 |
+
"step": 1300
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 0.44,
|
450 |
+
"learning_rate": 9.840848806366047e-05,
|
451 |
+
"loss": 1.8469,
|
452 |
+
"step": 1320
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 0.44,
|
456 |
+
"learning_rate": 9.575596816976126e-05,
|
457 |
+
"loss": 1.8405,
|
458 |
+
"step": 1340
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 0.45,
|
462 |
+
"learning_rate": 9.310344827586206e-05,
|
463 |
+
"loss": 1.8193,
|
464 |
+
"step": 1360
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.46,
|
468 |
+
"learning_rate": 9.045092838196286e-05,
|
469 |
+
"loss": 1.7731,
|
470 |
+
"step": 1380
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 0.46,
|
474 |
+
"learning_rate": 8.779840848806366e-05,
|
475 |
+
"loss": 1.8329,
|
476 |
+
"step": 1400
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 0.46,
|
480 |
+
"eval_loss": 1.9736511707305908,
|
481 |
+
"eval_runtime": 11956.5916,
|
482 |
+
"eval_samples_per_second": 1.118,
|
483 |
+
"eval_steps_per_second": 0.14,
|
484 |
+
"step": 1400
|
485 |
+
},
|
486 |
+
{
|
487 |
+
"epoch": 0.47,
|
488 |
+
"learning_rate": 8.514588859416444e-05,
|
489 |
+
"loss": 1.8872,
|
490 |
+
"step": 1420
|
491 |
+
},
|
492 |
+
{
|
493 |
+
"epoch": 0.48,
|
494 |
+
"learning_rate": 8.249336870026524e-05,
|
495 |
+
"loss": 1.8511,
|
496 |
+
"step": 1440
|
497 |
+
},
|
498 |
+
{
|
499 |
+
"epoch": 0.48,
|
500 |
+
"learning_rate": 7.984084880636604e-05,
|
501 |
+
"loss": 1.7523,
|
502 |
+
"step": 1460
|
503 |
+
},
|
504 |
+
{
|
505 |
+
"epoch": 0.49,
|
506 |
+
"learning_rate": 7.718832891246684e-05,
|
507 |
+
"loss": 1.8619,
|
508 |
+
"step": 1480
|
509 |
+
},
|
510 |
+
{
|
511 |
+
"epoch": 0.5,
|
512 |
+
"learning_rate": 7.453580901856764e-05,
|
513 |
+
"loss": 1.8363,
|
514 |
+
"step": 1500
|
515 |
+
},
|
516 |
+
{
|
517 |
+
"epoch": 0.5,
|
518 |
+
"learning_rate": 7.188328912466844e-05,
|
519 |
+
"loss": 1.7745,
|
520 |
+
"step": 1520
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.51,
|
524 |
+
"learning_rate": 6.923076923076922e-05,
|
525 |
+
"loss": 1.8164,
|
526 |
+
"step": 1540
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 0.52,
|
530 |
+
"learning_rate": 6.657824933687002e-05,
|
531 |
+
"loss": 1.7997,
|
532 |
+
"step": 1560
|
533 |
+
},
|
534 |
+
{
|
535 |
+
"epoch": 0.52,
|
536 |
+
"learning_rate": 6.392572944297081e-05,
|
537 |
+
"loss": 1.8377,
|
538 |
+
"step": 1580
|
539 |
+
},
|
540 |
+
{
|
541 |
+
"epoch": 0.53,
|
542 |
+
"learning_rate": 6.127320954907161e-05,
|
543 |
+
"loss": 1.851,
|
544 |
+
"step": 1600
|
545 |
+
},
|
546 |
+
{
|
547 |
+
"epoch": 0.53,
|
548 |
+
"eval_loss": 1.9729183912277222,
|
549 |
+
"eval_runtime": 11970.5767,
|
550 |
+
"eval_samples_per_second": 1.117,
|
551 |
+
"eval_steps_per_second": 0.14,
|
552 |
+
"step": 1600
|
553 |
+
},
|
554 |
+
{
|
555 |
+
"epoch": 0.54,
|
556 |
+
"learning_rate": 5.862068965517241e-05,
|
557 |
+
"loss": 1.8382,
|
558 |
+
"step": 1620
|
559 |
+
},
|
560 |
+
{
|
561 |
+
"epoch": 0.54,
|
562 |
+
"learning_rate": 5.59681697612732e-05,
|
563 |
+
"loss": 1.8863,
|
564 |
+
"step": 1640
|
565 |
+
},
|
566 |
+
{
|
567 |
+
"epoch": 0.55,
|
568 |
+
"learning_rate": 5.3315649867374e-05,
|
569 |
+
"loss": 1.762,
|
570 |
+
"step": 1660
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 0.56,
|
574 |
+
"learning_rate": 5.06631299734748e-05,
|
575 |
+
"loss": 1.822,
|
576 |
+
"step": 1680
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.56,
|
580 |
+
"learning_rate": 4.801061007957559e-05,
|
581 |
+
"loss": 1.7776,
|
582 |
+
"step": 1700
|
583 |
+
},
|
584 |
+
{
|
585 |
+
"epoch": 0.57,
|
586 |
+
"learning_rate": 4.5358090185676386e-05,
|
587 |
+
"loss": 1.809,
|
588 |
+
"step": 1720
|
589 |
+
},
|
590 |
+
{
|
591 |
+
"epoch": 0.58,
|
592 |
+
"learning_rate": 4.2705570291777186e-05,
|
593 |
+
"loss": 1.912,
|
594 |
+
"step": 1740
|
595 |
+
},
|
596 |
+
{
|
597 |
+
"epoch": 0.58,
|
598 |
+
"learning_rate": 4.005305039787798e-05,
|
599 |
+
"loss": 1.7557,
|
600 |
+
"step": 1760
|
601 |
+
},
|
602 |
+
{
|
603 |
+
"epoch": 0.59,
|
604 |
+
"learning_rate": 3.7400530503978774e-05,
|
605 |
+
"loss": 1.8232,
|
606 |
+
"step": 1780
|
607 |
+
},
|
608 |
+
{
|
609 |
+
"epoch": 0.6,
|
610 |
+
"learning_rate": 3.4748010610079574e-05,
|
611 |
+
"loss": 1.8613,
|
612 |
+
"step": 1800
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 0.6,
|
616 |
+
"eval_loss": 1.9704504013061523,
|
617 |
+
"eval_runtime": 11980.5334,
|
618 |
+
"eval_samples_per_second": 1.116,
|
619 |
+
"eval_steps_per_second": 0.139,
|
620 |
+
"step": 1800
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 0.6,
|
624 |
+
"learning_rate": 3.209549071618037e-05,
|
625 |
+
"loss": 1.8378,
|
626 |
+
"step": 1820
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 0.61,
|
630 |
+
"learning_rate": 2.9442970822281164e-05,
|
631 |
+
"loss": 1.8871,
|
632 |
+
"step": 1840
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.62,
|
636 |
+
"learning_rate": 2.679045092838196e-05,
|
637 |
+
"loss": 1.8821,
|
638 |
+
"step": 1860
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 0.62,
|
642 |
+
"learning_rate": 2.4137931034482755e-05,
|
643 |
+
"loss": 1.8487,
|
644 |
+
"step": 1880
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 0.63,
|
648 |
+
"learning_rate": 2.1485411140583555e-05,
|
649 |
+
"loss": 1.8439,
|
650 |
+
"step": 1900
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 0.64,
|
654 |
+
"learning_rate": 1.883289124668435e-05,
|
655 |
+
"loss": 1.8469,
|
656 |
+
"step": 1920
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 0.64,
|
660 |
+
"learning_rate": 1.6180371352785142e-05,
|
661 |
+
"loss": 1.7797,
|
662 |
+
"step": 1940
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 0.65,
|
666 |
+
"learning_rate": 1.352785145888594e-05,
|
667 |
+
"loss": 1.7795,
|
668 |
+
"step": 1960
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 0.66,
|
672 |
+
"learning_rate": 1.0875331564986736e-05,
|
673 |
+
"loss": 1.8483,
|
674 |
+
"step": 1980
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.66,
|
678 |
+
"learning_rate": 8.222811671087533e-06,
|
679 |
+
"loss": 1.8321,
|
680 |
+
"step": 2000
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 0.66,
|
684 |
+
"eval_loss": 1.9704707860946655,
|
685 |
+
"eval_runtime": 11964.4776,
|
686 |
+
"eval_samples_per_second": 1.117,
|
687 |
+
"eval_steps_per_second": 0.14,
|
688 |
+
"step": 2000
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.67,
|
692 |
+
"learning_rate": 5.5702917771883284e-06,
|
693 |
+
"loss": 1.7708,
|
694 |
+
"step": 2020
|
695 |
+
},
|
696 |
+
{
|
697 |
+
"epoch": 0.68,
|
698 |
+
"learning_rate": 2.9177718832891245e-06,
|
699 |
+
"loss": 1.8731,
|
700 |
+
"step": 2040
|
701 |
+
},
|
702 |
+
{
|
703 |
+
"epoch": 0.68,
|
704 |
+
"learning_rate": 2.652519893899204e-07,
|
705 |
+
"loss": 1.8181,
|
706 |
+
"step": 2060
|
707 |
+
},
|
708 |
+
{
|
709 |
+
"epoch": 0.69,
|
710 |
+
"learning_rate": 0.0,
|
711 |
+
"loss": 1.8692,
|
712 |
+
"step": 2080
|
713 |
+
},
|
714 |
+
{
|
715 |
+
"epoch": 0.7,
|
716 |
+
"learning_rate": 0.0,
|
717 |
+
"loss": 1.8602,
|
718 |
+
"step": 2100
|
719 |
+
},
|
720 |
+
{
|
721 |
+
"epoch": 0.7,
|
722 |
+
"learning_rate": 0.0,
|
723 |
+
"loss": 1.8891,
|
724 |
+
"step": 2120
|
725 |
+
},
|
726 |
+
{
|
727 |
+
"epoch": 0.71,
|
728 |
+
"learning_rate": 0.0,
|
729 |
+
"loss": 1.8205,
|
730 |
+
"step": 2140
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.72,
|
734 |
+
"learning_rate": 0.0,
|
735 |
+
"loss": 1.8242,
|
736 |
+
"step": 2160
|
737 |
+
},
|
738 |
+
{
|
739 |
+
"epoch": 0.72,
|
740 |
+
"learning_rate": 0.0,
|
741 |
+
"loss": 1.8661,
|
742 |
+
"step": 2180
|
743 |
+
},
|
744 |
+
{
|
745 |
+
"epoch": 0.73,
|
746 |
+
"learning_rate": 0.0,
|
747 |
+
"loss": 1.7982,
|
748 |
+
"step": 2200
|
749 |
+
},
|
750 |
+
{
|
751 |
+
"epoch": 0.73,
|
752 |
+
"eval_loss": 1.970503568649292,
|
753 |
+
"eval_runtime": 11950.1314,
|
754 |
+
"eval_samples_per_second": 1.119,
|
755 |
+
"eval_steps_per_second": 0.14,
|
756 |
+
"step": 2200
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 0.74,
|
760 |
+
"learning_rate": 0.0,
|
761 |
+
"loss": 1.7967,
|
762 |
+
"step": 2220
|
763 |
+
},
|
764 |
+
{
|
765 |
+
"epoch": 0.74,
|
766 |
+
"learning_rate": 0.0,
|
767 |
+
"loss": 1.8346,
|
768 |
+
"step": 2240
|
769 |
+
},
|
770 |
+
{
|
771 |
+
"epoch": 0.75,
|
772 |
+
"learning_rate": 0.0,
|
773 |
+
"loss": 1.8242,
|
774 |
+
"step": 2260
|
775 |
+
},
|
776 |
+
{
|
777 |
+
"epoch": 0.76,
|
778 |
+
"learning_rate": 0.0,
|
779 |
+
"loss": 1.8429,
|
780 |
+
"step": 2280
|
781 |
+
},
|
782 |
+
{
|
783 |
+
"epoch": 0.76,
|
784 |
+
"learning_rate": 0.0,
|
785 |
+
"loss": 1.8795,
|
786 |
+
"step": 2300
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.77,
|
790 |
+
"learning_rate": 0.0,
|
791 |
+
"loss": 1.8967,
|
792 |
+
"step": 2320
|
793 |
+
},
|
794 |
+
{
|
795 |
+
"epoch": 0.78,
|
796 |
+
"learning_rate": 0.0,
|
797 |
+
"loss": 1.815,
|
798 |
+
"step": 2340
|
799 |
+
},
|
800 |
+
{
|
801 |
+
"epoch": 0.78,
|
802 |
+
"learning_rate": 0.0,
|
803 |
+
"loss": 1.8165,
|
804 |
+
"step": 2360
|
805 |
+
},
|
806 |
+
{
|
807 |
+
"epoch": 0.79,
|
808 |
+
"learning_rate": 0.0,
|
809 |
+
"loss": 1.838,
|
810 |
+
"step": 2380
|
811 |
+
},
|
812 |
+
{
|
813 |
+
"epoch": 0.8,
|
814 |
+
"learning_rate": 0.0,
|
815 |
+
"loss": 1.7798,
|
816 |
+
"step": 2400
|
817 |
+
},
|
818 |
+
{
|
819 |
+
"epoch": 0.8,
|
820 |
+
"eval_loss": 1.970503568649292,
|
821 |
+
"eval_runtime": 11951.0888,
|
822 |
+
"eval_samples_per_second": 1.119,
|
823 |
+
"eval_steps_per_second": 0.14,
|
824 |
+
"step": 2400
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 0.8,
|
828 |
+
"learning_rate": 0.0,
|
829 |
+
"loss": 1.7999,
|
830 |
+
"step": 2420
|
831 |
+
},
|
832 |
+
{
|
833 |
+
"epoch": 0.81,
|
834 |
+
"learning_rate": 0.0,
|
835 |
+
"loss": 1.7957,
|
836 |
+
"step": 2440
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 0.82,
|
840 |
+
"learning_rate": 0.0,
|
841 |
+
"loss": 1.8306,
|
842 |
+
"step": 2460
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.82,
|
846 |
+
"learning_rate": 0.0,
|
847 |
+
"loss": 1.87,
|
848 |
+
"step": 2480
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 0.83,
|
852 |
+
"learning_rate": 0.0,
|
853 |
+
"loss": 1.8046,
|
854 |
+
"step": 2500
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 0.84,
|
858 |
+
"learning_rate": 0.0,
|
859 |
+
"loss": 1.76,
|
860 |
+
"step": 2520
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 0.84,
|
864 |
+
"learning_rate": 0.0,
|
865 |
+
"loss": 1.869,
|
866 |
+
"step": 2540
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 0.85,
|
870 |
+
"learning_rate": 0.0,
|
871 |
+
"loss": 1.8684,
|
872 |
+
"step": 2560
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 0.86,
|
876 |
+
"learning_rate": 0.0,
|
877 |
+
"loss": 1.8757,
|
878 |
+
"step": 2580
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 0.86,
|
882 |
+
"learning_rate": 0.0,
|
883 |
+
"loss": 1.868,
|
884 |
+
"step": 2600
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.86,
|
888 |
+
"eval_loss": 1.970503568649292,
|
889 |
+
"eval_runtime": 11952.488,
|
890 |
+
"eval_samples_per_second": 1.118,
|
891 |
+
"eval_steps_per_second": 0.14,
|
892 |
+
"step": 2600
|
893 |
+
},
|
894 |
+
{
|
895 |
+
"epoch": 0.87,
|
896 |
+
"learning_rate": 0.0,
|
897 |
+
"loss": 1.8228,
|
898 |
+
"step": 2620
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.88,
|
902 |
+
"learning_rate": 0.0,
|
903 |
+
"loss": 1.8721,
|
904 |
+
"step": 2640
|
905 |
+
},
|
906 |
+
{
|
907 |
+
"epoch": 0.88,
|
908 |
+
"learning_rate": 0.0,
|
909 |
+
"loss": 1.8557,
|
910 |
+
"step": 2660
|
911 |
+
},
|
912 |
+
{
|
913 |
+
"epoch": 0.89,
|
914 |
+
"learning_rate": 0.0,
|
915 |
+
"loss": 1.8545,
|
916 |
+
"step": 2680
|
917 |
+
},
|
918 |
+
{
|
919 |
+
"epoch": 0.9,
|
920 |
+
"learning_rate": 0.0,
|
921 |
+
"loss": 1.8366,
|
922 |
+
"step": 2700
|
923 |
+
},
|
924 |
+
{
|
925 |
+
"epoch": 0.9,
|
926 |
+
"learning_rate": 0.0,
|
927 |
+
"loss": 1.8381,
|
928 |
+
"step": 2720
|
929 |
+
},
|
930 |
+
{
|
931 |
+
"epoch": 0.91,
|
932 |
+
"learning_rate": 0.0,
|
933 |
+
"loss": 1.8305,
|
934 |
+
"step": 2740
|
935 |
+
},
|
936 |
+
{
|
937 |
+
"epoch": 0.92,
|
938 |
+
"learning_rate": 0.0,
|
939 |
+
"loss": 1.8162,
|
940 |
+
"step": 2760
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.92,
|
944 |
+
"learning_rate": 0.0,
|
945 |
+
"loss": 1.8109,
|
946 |
+
"step": 2780
|
947 |
+
},
|
948 |
+
{
|
949 |
+
"epoch": 0.93,
|
950 |
+
"learning_rate": 0.0,
|
951 |
+
"loss": 1.8694,
|
952 |
+
"step": 2800
|
953 |
+
},
|
954 |
+
{
|
955 |
+
"epoch": 0.93,
|
956 |
+
"eval_loss": 1.970503568649292,
|
957 |
+
"eval_runtime": 11969.9504,
|
958 |
+
"eval_samples_per_second": 1.117,
|
959 |
+
"eval_steps_per_second": 0.14,
|
960 |
+
"step": 2800
|
961 |
+
},
|
962 |
+
{
|
963 |
+
"epoch": 0.94,
|
964 |
+
"learning_rate": 0.0,
|
965 |
+
"loss": 1.8201,
|
966 |
+
"step": 2820
|
967 |
+
},
|
968 |
+
{
|
969 |
+
"epoch": 0.94,
|
970 |
+
"learning_rate": 0.0,
|
971 |
+
"loss": 1.848,
|
972 |
+
"step": 2840
|
973 |
+
},
|
974 |
+
{
|
975 |
+
"epoch": 0.95,
|
976 |
+
"learning_rate": 0.0,
|
977 |
+
"loss": 1.8138,
|
978 |
+
"step": 2860
|
979 |
+
},
|
980 |
+
{
|
981 |
+
"epoch": 0.96,
|
982 |
+
"learning_rate": 0.0,
|
983 |
+
"loss": 1.8554,
|
984 |
+
"step": 2880
|
985 |
+
},
|
986 |
+
{
|
987 |
+
"epoch": 0.96,
|
988 |
+
"learning_rate": 0.0,
|
989 |
+
"loss": 1.8321,
|
990 |
+
"step": 2900
|
991 |
+
},
|
992 |
+
{
|
993 |
+
"epoch": 0.97,
|
994 |
+
"learning_rate": 0.0,
|
995 |
+
"loss": 1.8404,
|
996 |
+
"step": 2920
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.98,
|
1000 |
+
"learning_rate": 0.0,
|
1001 |
+
"loss": 1.7883,
|
1002 |
+
"step": 2940
|
1003 |
+
},
|
1004 |
+
{
|
1005 |
+
"epoch": 0.98,
|
1006 |
+
"learning_rate": 0.0,
|
1007 |
+
"loss": 1.8733,
|
1008 |
+
"step": 2960
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 0.99,
|
1012 |
+
"learning_rate": 0.0,
|
1013 |
+
"loss": 1.8074,
|
1014 |
+
"step": 2980
|
1015 |
+
},
|
1016 |
+
{
|
1017 |
+
"epoch": 1.0,
|
1018 |
+
"learning_rate": 0.0,
|
1019 |
+
"loss": 1.887,
|
1020 |
+
"step": 3000
|
1021 |
+
},
|
1022 |
+
{
|
1023 |
+
"epoch": 1.0,
|
1024 |
+
"eval_loss": 1.970503568649292,
|
1025 |
+
"eval_runtime": 11974.5637,
|
1026 |
+
"eval_samples_per_second": 1.116,
|
1027 |
+
"eval_steps_per_second": 0.14,
|
1028 |
+
"step": 3000
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 1.0,
|
1032 |
+
"learning_rate": 0.0,
|
1033 |
+
"loss": 1.7165,
|
1034 |
+
"step": 3020
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 1.01,
|
1038 |
+
"learning_rate": 0.0,
|
1039 |
+
"loss": 1.8041,
|
1040 |
+
"step": 3040
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 1.01,
|
1044 |
+
"learning_rate": 0.0,
|
1045 |
+
"loss": 1.8031,
|
1046 |
+
"step": 3060
|
1047 |
+
},
|
1048 |
+
{
|
1049 |
+
"epoch": 1.02,
|
1050 |
+
"learning_rate": 0.0,
|
1051 |
+
"loss": 1.8493,
|
1052 |
+
"step": 3080
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 1.03,
|
1056 |
+
"learning_rate": 0.0,
|
1057 |
+
"loss": 1.8475,
|
1058 |
+
"step": 3100
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 1.03,
|
1062 |
+
"learning_rate": 0.0,
|
1063 |
+
"loss": 1.7994,
|
1064 |
+
"step": 3120
|
1065 |
+
},
|
1066 |
+
{
|
1067 |
+
"epoch": 1.04,
|
1068 |
+
"learning_rate": 0.0,
|
1069 |
+
"loss": 1.7904,
|
1070 |
+
"step": 3140
|
1071 |
+
},
|
1072 |
+
{
|
1073 |
+
"epoch": 1.05,
|
1074 |
+
"learning_rate": 0.0,
|
1075 |
+
"loss": 1.8078,
|
1076 |
+
"step": 3160
|
1077 |
+
},
|
1078 |
+
{
|
1079 |
+
"epoch": 1.05,
|
1080 |
+
"learning_rate": 0.0,
|
1081 |
+
"loss": 1.8625,
|
1082 |
+
"step": 3180
|
1083 |
+
},
|
1084 |
+
{
|
1085 |
+
"epoch": 1.06,
|
1086 |
+
"learning_rate": 0.0,
|
1087 |
+
"loss": 1.7874,
|
1088 |
+
"step": 3200
|
1089 |
+
},
|
1090 |
+
{
|
1091 |
+
"epoch": 1.06,
|
1092 |
+
"eval_loss": 1.970503568649292,
|
1093 |
+
"eval_runtime": 11966.566,
|
1094 |
+
"eval_samples_per_second": 1.117,
|
1095 |
+
"eval_steps_per_second": 0.14,
|
1096 |
+
"step": 3200
|
1097 |
+
},
|
1098 |
+
{
|
1099 |
+
"epoch": 1.07,
|
1100 |
+
"learning_rate": 0.0,
|
1101 |
+
"loss": 1.7956,
|
1102 |
+
"step": 3220
|
1103 |
+
},
|
1104 |
+
{
|
1105 |
+
"epoch": 1.07,
|
1106 |
+
"learning_rate": 0.0,
|
1107 |
+
"loss": 1.7517,
|
1108 |
+
"step": 3240
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 1.08,
|
1112 |
+
"learning_rate": 0.0,
|
1113 |
+
"loss": 1.7861,
|
1114 |
+
"step": 3260
|
1115 |
+
},
|
1116 |
+
{
|
1117 |
+
"epoch": 1.09,
|
1118 |
+
"learning_rate": 0.0,
|
1119 |
+
"loss": 1.7957,
|
1120 |
+
"step": 3280
|
1121 |
+
},
|
1122 |
+
{
|
1123 |
+
"epoch": 1.09,
|
1124 |
+
"learning_rate": 0.0,
|
1125 |
+
"loss": 1.8147,
|
1126 |
+
"step": 3300
|
1127 |
+
},
|
1128 |
+
{
|
1129 |
+
"epoch": 1.1,
|
1130 |
+
"learning_rate": 0.0,
|
1131 |
+
"loss": 1.7795,
|
1132 |
+
"step": 3320
|
1133 |
+
},
|
1134 |
+
{
|
1135 |
+
"epoch": 1.11,
|
1136 |
+
"learning_rate": 0.0,
|
1137 |
+
"loss": 1.7532,
|
1138 |
+
"step": 3340
|
1139 |
+
},
|
1140 |
+
{
|
1141 |
+
"epoch": 1.11,
|
1142 |
+
"learning_rate": 0.0,
|
1143 |
+
"loss": 1.8414,
|
1144 |
+
"step": 3360
|
1145 |
+
},
|
1146 |
+
{
|
1147 |
+
"epoch": 1.12,
|
1148 |
+
"learning_rate": 0.0,
|
1149 |
+
"loss": 1.771,
|
1150 |
+
"step": 3380
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 1.13,
|
1154 |
+
"learning_rate": 0.0,
|
1155 |
+
"loss": 1.8119,
|
1156 |
+
"step": 3400
|
1157 |
+
},
|
1158 |
+
{
|
1159 |
+
"epoch": 1.13,
|
1160 |
+
"eval_loss": 1.970503568649292,
|
1161 |
+
"eval_runtime": 11954.7549,
|
1162 |
+
"eval_samples_per_second": 1.118,
|
1163 |
+
"eval_steps_per_second": 0.14,
|
1164 |
+
"step": 3400
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 1.13,
|
1168 |
+
"learning_rate": 0.0,
|
1169 |
+
"loss": 1.7944,
|
1170 |
+
"step": 3420
|
1171 |
+
},
|
1172 |
+
{
|
1173 |
+
"epoch": 1.14,
|
1174 |
+
"learning_rate": 0.0,
|
1175 |
+
"loss": 1.7855,
|
1176 |
+
"step": 3440
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 1.15,
|
1180 |
+
"learning_rate": 0.0,
|
1181 |
+
"loss": 1.8574,
|
1182 |
+
"step": 3460
|
1183 |
+
},
|
1184 |
+
{
|
1185 |
+
"epoch": 1.15,
|
1186 |
+
"learning_rate": 0.0,
|
1187 |
+
"loss": 1.8189,
|
1188 |
+
"step": 3480
|
1189 |
+
},
|
1190 |
+
{
|
1191 |
+
"epoch": 1.16,
|
1192 |
+
"learning_rate": 0.0,
|
1193 |
+
"loss": 1.8033,
|
1194 |
+
"step": 3500
|
1195 |
+
},
|
1196 |
+
{
|
1197 |
+
"epoch": 1.17,
|
1198 |
+
"learning_rate": 0.0,
|
1199 |
+
"loss": 1.7689,
|
1200 |
+
"step": 3520
|
1201 |
+
},
|
1202 |
+
{
|
1203 |
+
"epoch": 1.17,
|
1204 |
+
"learning_rate": 0.0,
|
1205 |
+
"loss": 1.8455,
|
1206 |
+
"step": 3540
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 1.18,
|
1210 |
+
"learning_rate": 0.0,
|
1211 |
+
"loss": 1.8255,
|
1212 |
+
"step": 3560
|
1213 |
+
},
|
1214 |
+
{
|
1215 |
+
"epoch": 1.19,
|
1216 |
+
"learning_rate": 0.0,
|
1217 |
+
"loss": 1.8431,
|
1218 |
+
"step": 3580
|
1219 |
+
},
|
1220 |
+
{
|
1221 |
+
"epoch": 1.19,
|
1222 |
+
"learning_rate": 0.0,
|
1223 |
+
"loss": 1.7924,
|
1224 |
+
"step": 3600
|
1225 |
+
},
|
1226 |
+
{
|
1227 |
+
"epoch": 1.19,
|
1228 |
+
"eval_loss": 1.970503568649292,
|
1229 |
+
"eval_runtime": 11953.5885,
|
1230 |
+
"eval_samples_per_second": 1.118,
|
1231 |
+
"eval_steps_per_second": 0.14,
|
1232 |
+
"step": 3600
|
1233 |
+
},
|
1234 |
+
{
|
1235 |
+
"epoch": 1.2,
|
1236 |
+
"learning_rate": 0.0,
|
1237 |
+
"loss": 1.8475,
|
1238 |
+
"step": 3620
|
1239 |
+
},
|
1240 |
+
{
|
1241 |
+
"epoch": 1.21,
|
1242 |
+
"learning_rate": 0.0,
|
1243 |
+
"loss": 1.6746,
|
1244 |
+
"step": 3640
|
1245 |
+
},
|
1246 |
+
{
|
1247 |
+
"epoch": 1.21,
|
1248 |
+
"learning_rate": 0.0,
|
1249 |
+
"loss": 1.8006,
|
1250 |
+
"step": 3660
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 1.22,
|
1254 |
+
"learning_rate": 0.0,
|
1255 |
+
"loss": 1.816,
|
1256 |
+
"step": 3680
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 1.23,
|
1260 |
+
"learning_rate": 0.0,
|
1261 |
+
"loss": 1.7486,
|
1262 |
+
"step": 3700
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 1.23,
|
1266 |
+
"learning_rate": 0.0,
|
1267 |
+
"loss": 1.8086,
|
1268 |
+
"step": 3720
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 1.24,
|
1272 |
+
"learning_rate": 0.0,
|
1273 |
+
"loss": 1.8241,
|
1274 |
+
"step": 3740
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 1.25,
|
1278 |
+
"learning_rate": 0.0,
|
1279 |
+
"loss": 1.823,
|
1280 |
+
"step": 3760
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 1.25,
|
1284 |
+
"learning_rate": 0.0,
|
1285 |
+
"loss": 1.7365,
|
1286 |
+
"step": 3780
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 1.26,
|
1290 |
+
"learning_rate": 0.0,
|
1291 |
+
"loss": 1.801,
|
1292 |
+
"step": 3800
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 1.26,
|
1296 |
+
"eval_loss": 1.970503568649292,
|
1297 |
+
"eval_runtime": 11952.013,
|
1298 |
+
"eval_samples_per_second": 1.118,
|
1299 |
+
"eval_steps_per_second": 0.14,
|
1300 |
+
"step": 3800
|
1301 |
+
},
|
1302 |
+
{
|
1303 |
+
"epoch": 1.27,
|
1304 |
+
"learning_rate": 0.0,
|
1305 |
+
"loss": 1.8497,
|
1306 |
+
"step": 3820
|
1307 |
+
},
|
1308 |
+
{
|
1309 |
+
"epoch": 1.27,
|
1310 |
+
"learning_rate": 0.0,
|
1311 |
+
"loss": 1.7803,
|
1312 |
+
"step": 3840
|
1313 |
+
},
|
1314 |
+
{
|
1315 |
+
"epoch": 1.28,
|
1316 |
+
"learning_rate": 0.0,
|
1317 |
+
"loss": 1.6921,
|
1318 |
+
"step": 3860
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 1.29,
|
1322 |
+
"learning_rate": 0.0,
|
1323 |
+
"loss": 1.8508,
|
1324 |
+
"step": 3880
|
1325 |
+
},
|
1326 |
+
{
|
1327 |
+
"epoch": 1.29,
|
1328 |
+
"learning_rate": 0.0,
|
1329 |
+
"loss": 1.8816,
|
1330 |
+
"step": 3900
|
1331 |
+
},
|
1332 |
+
{
|
1333 |
+
"epoch": 1.3,
|
1334 |
+
"learning_rate": 0.0,
|
1335 |
+
"loss": 1.8105,
|
1336 |
+
"step": 3920
|
1337 |
+
},
|
1338 |
+
{
|
1339 |
+
"epoch": 1.31,
|
1340 |
+
"learning_rate": 0.0,
|
1341 |
+
"loss": 1.8477,
|
1342 |
+
"step": 3940
|
1343 |
+
},
|
1344 |
+
{
|
1345 |
+
"epoch": 1.31,
|
1346 |
+
"learning_rate": 0.0,
|
1347 |
+
"loss": 1.841,
|
1348 |
+
"step": 3960
|
1349 |
+
},
|
1350 |
+
{
|
1351 |
+
"epoch": 1.32,
|
1352 |
+
"learning_rate": 0.0,
|
1353 |
+
"loss": 1.7599,
|
1354 |
+
"step": 3980
|
1355 |
+
},
|
1356 |
+
{
|
1357 |
+
"epoch": 1.33,
|
1358 |
+
"learning_rate": 0.0,
|
1359 |
+
"loss": 1.7074,
|
1360 |
+
"step": 4000
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 1.33,
|
1364 |
+
"eval_loss": 1.970503568649292,
|
1365 |
+
"eval_runtime": 11952.3623,
|
1366 |
+
"eval_samples_per_second": 1.118,
|
1367 |
+
"eval_steps_per_second": 0.14,
|
1368 |
+
"step": 4000
|
1369 |
+
},
|
1370 |
+
{
|
1371 |
+
"epoch": 1.33,
|
1372 |
+
"learning_rate": 0.0,
|
1373 |
+
"loss": 1.7529,
|
1374 |
+
"step": 4020
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 1.34,
|
1378 |
+
"learning_rate": 0.0,
|
1379 |
+
"loss": 1.8928,
|
1380 |
+
"step": 4040
|
1381 |
+
},
|
1382 |
+
{
|
1383 |
+
"epoch": 1.35,
|
1384 |
+
"learning_rate": 0.0,
|
1385 |
+
"loss": 1.8585,
|
1386 |
+
"step": 4060
|
1387 |
+
},
|
1388 |
+
{
|
1389 |
+
"epoch": 1.35,
|
1390 |
+
"learning_rate": 0.0,
|
1391 |
+
"loss": 1.8279,
|
1392 |
+
"step": 4080
|
1393 |
+
},
|
1394 |
+
{
|
1395 |
+
"epoch": 1.36,
|
1396 |
+
"learning_rate": 0.0,
|
1397 |
+
"loss": 1.7949,
|
1398 |
+
"step": 4100
|
1399 |
+
},
|
1400 |
+
{
|
1401 |
+
"epoch": 1.37,
|
1402 |
+
"learning_rate": 0.0,
|
1403 |
+
"loss": 1.8488,
|
1404 |
+
"step": 4120
|
1405 |
+
},
|
1406 |
+
{
|
1407 |
+
"epoch": 1.37,
|
1408 |
+
"learning_rate": 0.0,
|
1409 |
+
"loss": 1.7572,
|
1410 |
+
"step": 4140
|
1411 |
+
},
|
1412 |
+
{
|
1413 |
+
"epoch": 1.38,
|
1414 |
+
"learning_rate": 0.0,
|
1415 |
+
"loss": 1.7533,
|
1416 |
+
"step": 4160
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 1.39,
|
1420 |
+
"learning_rate": 0.0,
|
1421 |
+
"loss": 1.772,
|
1422 |
+
"step": 4180
|
1423 |
+
},
|
1424 |
+
{
|
1425 |
+
"epoch": 1.39,
|
1426 |
+
"learning_rate": 0.0,
|
1427 |
+
"loss": 1.8338,
|
1428 |
+
"step": 4200
|
1429 |
+
},
|
1430 |
+
{
|
1431 |
+
"epoch": 1.39,
|
1432 |
+
"eval_loss": 1.970503568649292,
|
1433 |
+
"eval_runtime": 11954.2113,
|
1434 |
+
"eval_samples_per_second": 1.118,
|
1435 |
+
"eval_steps_per_second": 0.14,
|
1436 |
+
"step": 4200
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 1.4,
|
1440 |
+
"learning_rate": 0.0,
|
1441 |
+
"loss": 1.8084,
|
1442 |
+
"step": 4220
|
1443 |
+
},
|
1444 |
+
{
|
1445 |
+
"epoch": 1.41,
|
1446 |
+
"learning_rate": 0.0,
|
1447 |
+
"loss": 1.8791,
|
1448 |
+
"step": 4240
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 1.41,
|
1452 |
+
"learning_rate": 0.0,
|
1453 |
+
"loss": 1.7906,
|
1454 |
+
"step": 4260
|
1455 |
+
},
|
1456 |
+
{
|
1457 |
+
"epoch": 1.42,
|
1458 |
+
"learning_rate": 0.0,
|
1459 |
+
"loss": 1.8669,
|
1460 |
+
"step": 4280
|
1461 |
+
},
|
1462 |
+
{
|
1463 |
+
"epoch": 1.43,
|
1464 |
+
"learning_rate": 0.0,
|
1465 |
+
"loss": 1.8108,
|
1466 |
+
"step": 4300
|
1467 |
+
},
|
1468 |
+
{
|
1469 |
+
"epoch": 1.43,
|
1470 |
+
"learning_rate": 0.0,
|
1471 |
+
"loss": 1.7769,
|
1472 |
+
"step": 4320
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 1.44,
|
1476 |
+
"learning_rate": 0.0,
|
1477 |
+
"loss": 1.7599,
|
1478 |
+
"step": 4340
|
1479 |
+
},
|
1480 |
+
{
|
1481 |
+
"epoch": 1.45,
|
1482 |
+
"learning_rate": 0.0,
|
1483 |
+
"loss": 1.843,
|
1484 |
+
"step": 4360
|
1485 |
+
},
|
1486 |
+
{
|
1487 |
+
"epoch": 1.45,
|
1488 |
+
"learning_rate": 0.0,
|
1489 |
+
"loss": 1.8608,
|
1490 |
+
"step": 4380
|
1491 |
+
},
|
1492 |
+
{
|
1493 |
+
"epoch": 1.46,
|
1494 |
+
"learning_rate": 0.0,
|
1495 |
+
"loss": 1.8382,
|
1496 |
+
"step": 4400
|
1497 |
+
},
|
1498 |
+
{
|
1499 |
+
"epoch": 1.46,
|
1500 |
+
"eval_loss": 1.970503568649292,
|
1501 |
+
"eval_runtime": 11958.7166,
|
1502 |
+
"eval_samples_per_second": 1.118,
|
1503 |
+
"eval_steps_per_second": 0.14,
|
1504 |
+
"step": 4400
|
1505 |
+
},
|
1506 |
+
{
|
1507 |
+
"epoch": 1.47,
|
1508 |
+
"learning_rate": 0.0,
|
1509 |
+
"loss": 1.8002,
|
1510 |
+
"step": 4420
|
1511 |
+
},
|
1512 |
+
{
|
1513 |
+
"epoch": 1.47,
|
1514 |
+
"learning_rate": 0.0,
|
1515 |
+
"loss": 1.8292,
|
1516 |
+
"step": 4440
|
1517 |
+
},
|
1518 |
+
{
|
1519 |
+
"epoch": 1.48,
|
1520 |
+
"learning_rate": 0.0,
|
1521 |
+
"loss": 1.8184,
|
1522 |
+
"step": 4460
|
1523 |
+
},
|
1524 |
+
{
|
1525 |
+
"epoch": 1.49,
|
1526 |
+
"learning_rate": 0.0,
|
1527 |
+
"loss": 1.7712,
|
1528 |
+
"step": 4480
|
1529 |
+
},
|
1530 |
+
{
|
1531 |
+
"epoch": 1.49,
|
1532 |
+
"learning_rate": 0.0,
|
1533 |
+
"loss": 1.8307,
|
1534 |
+
"step": 4500
|
1535 |
+
},
|
1536 |
+
{
|
1537 |
+
"epoch": 1.5,
|
1538 |
+
"learning_rate": 0.0,
|
1539 |
+
"loss": 1.7883,
|
1540 |
+
"step": 4520
|
1541 |
+
},
|
1542 |
+
{
|
1543 |
+
"epoch": 1.51,
|
1544 |
+
"learning_rate": 0.0,
|
1545 |
+
"loss": 1.8106,
|
1546 |
+
"step": 4540
|
1547 |
+
},
|
1548 |
+
{
|
1549 |
+
"epoch": 1.51,
|
1550 |
+
"learning_rate": 0.0,
|
1551 |
+
"loss": 1.7999,
|
1552 |
+
"step": 4560
|
1553 |
+
},
|
1554 |
+
{
|
1555 |
+
"epoch": 1.52,
|
1556 |
+
"learning_rate": 0.0,
|
1557 |
+
"loss": 1.8353,
|
1558 |
+
"step": 4580
|
1559 |
+
},
|
1560 |
+
{
|
1561 |
+
"epoch": 1.53,
|
1562 |
+
"learning_rate": 0.0,
|
1563 |
+
"loss": 1.8366,
|
1564 |
+
"step": 4600
|
1565 |
+
},
|
1566 |
+
{
|
1567 |
+
"epoch": 1.53,
|
1568 |
+
"eval_loss": 1.970503568649292,
|
1569 |
+
"eval_runtime": 11971.0491,
|
1570 |
+
"eval_samples_per_second": 1.117,
|
1571 |
+
"eval_steps_per_second": 0.14,
|
1572 |
+
"step": 4600
|
1573 |
+
},
|
1574 |
+
{
|
1575 |
+
"epoch": 1.53,
|
1576 |
+
"learning_rate": 0.0,
|
1577 |
+
"loss": 1.8018,
|
1578 |
+
"step": 4620
|
1579 |
+
},
|
1580 |
+
{
|
1581 |
+
"epoch": 1.54,
|
1582 |
+
"learning_rate": 0.0,
|
1583 |
+
"loss": 1.84,
|
1584 |
+
"step": 4640
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 1.55,
|
1588 |
+
"learning_rate": 0.0,
|
1589 |
+
"loss": 1.7784,
|
1590 |
+
"step": 4660
|
1591 |
+
},
|
1592 |
+
{
|
1593 |
+
"epoch": 1.55,
|
1594 |
+
"learning_rate": 0.0,
|
1595 |
+
"loss": 1.794,
|
1596 |
+
"step": 4680
|
1597 |
+
},
|
1598 |
+
{
|
1599 |
+
"epoch": 1.56,
|
1600 |
+
"learning_rate": 0.0,
|
1601 |
+
"loss": 1.8237,
|
1602 |
+
"step": 4700
|
1603 |
+
},
|
1604 |
+
{
|
1605 |
+
"epoch": 1.57,
|
1606 |
+
"learning_rate": 0.0,
|
1607 |
+
"loss": 1.7697,
|
1608 |
+
"step": 4720
|
1609 |
+
},
|
1610 |
+
{
|
1611 |
+
"epoch": 1.57,
|
1612 |
+
"learning_rate": 0.0,
|
1613 |
+
"loss": 1.8482,
|
1614 |
+
"step": 4740
|
1615 |
+
},
|
1616 |
+
{
|
1617 |
+
"epoch": 1.58,
|
1618 |
+
"learning_rate": 0.0,
|
1619 |
+
"loss": 1.8008,
|
1620 |
+
"step": 4760
|
1621 |
+
},
|
1622 |
+
{
|
1623 |
+
"epoch": 1.59,
|
1624 |
+
"learning_rate": 0.0,
|
1625 |
+
"loss": 1.8082,
|
1626 |
+
"step": 4780
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 1.59,
|
1630 |
+
"learning_rate": 0.0,
|
1631 |
+
"loss": 1.7799,
|
1632 |
+
"step": 4800
|
1633 |
+
},
|
1634 |
+
{
|
1635 |
+
"epoch": 1.59,
|
1636 |
+
"eval_loss": 1.970503568649292,
|
1637 |
+
"eval_runtime": 11955.9821,
|
1638 |
+
"eval_samples_per_second": 1.118,
|
1639 |
+
"eval_steps_per_second": 0.14,
|
1640 |
+
"step": 4800
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 1.6,
|
1644 |
+
"learning_rate": 0.0,
|
1645 |
+
"loss": 1.8339,
|
1646 |
+
"step": 4820
|
1647 |
+
},
|
1648 |
+
{
|
1649 |
+
"epoch": 1.61,
|
1650 |
+
"learning_rate": 0.0,
|
1651 |
+
"loss": 1.8072,
|
1652 |
+
"step": 4840
|
1653 |
+
},
|
1654 |
+
{
|
1655 |
+
"epoch": 1.61,
|
1656 |
+
"learning_rate": 0.0,
|
1657 |
+
"loss": 1.8024,
|
1658 |
+
"step": 4860
|
1659 |
+
},
|
1660 |
+
{
|
1661 |
+
"epoch": 1.62,
|
1662 |
+
"learning_rate": 0.0,
|
1663 |
+
"loss": 1.8609,
|
1664 |
+
"step": 4880
|
1665 |
+
},
|
1666 |
+
{
|
1667 |
+
"epoch": 1.63,
|
1668 |
+
"learning_rate": 0.0,
|
1669 |
+
"loss": 1.8769,
|
1670 |
+
"step": 4900
|
1671 |
+
},
|
1672 |
+
{
|
1673 |
+
"epoch": 1.63,
|
1674 |
+
"learning_rate": 0.0,
|
1675 |
+
"loss": 1.808,
|
1676 |
+
"step": 4920
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 1.64,
|
1680 |
+
"learning_rate": 0.0,
|
1681 |
+
"loss": 1.8482,
|
1682 |
+
"step": 4940
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 1.65,
|
1686 |
+
"learning_rate": 0.0,
|
1687 |
+
"loss": 1.8116,
|
1688 |
+
"step": 4960
|
1689 |
+
},
|
1690 |
+
{
|
1691 |
+
"epoch": 1.65,
|
1692 |
+
"learning_rate": 0.0,
|
1693 |
+
"loss": 1.7922,
|
1694 |
+
"step": 4980
|
1695 |
+
},
|
1696 |
+
{
|
1697 |
+
"epoch": 1.66,
|
1698 |
+
"learning_rate": 0.0,
|
1699 |
+
"loss": 1.8409,
|
1700 |
+
"step": 5000
|
1701 |
+
},
|
1702 |
+
{
|
1703 |
+
"epoch": 1.66,
|
1704 |
+
"eval_loss": 1.970503568649292,
|
1705 |
+
"eval_runtime": 11953.6568,
|
1706 |
+
"eval_samples_per_second": 1.118,
|
1707 |
+
"eval_steps_per_second": 0.14,
|
1708 |
+
"step": 5000
|
1709 |
+
},
|
1710 |
+
{
|
1711 |
+
"epoch": 1.67,
|
1712 |
+
"learning_rate": 0.0,
|
1713 |
+
"loss": 1.867,
|
1714 |
+
"step": 5020
|
1715 |
+
},
|
1716 |
+
{
|
1717 |
+
"epoch": 1.67,
|
1718 |
+
"learning_rate": 0.0,
|
1719 |
+
"loss": 1.8902,
|
1720 |
+
"step": 5040
|
1721 |
+
},
|
1722 |
+
{
|
1723 |
+
"epoch": 1.68,
|
1724 |
+
"learning_rate": 0.0,
|
1725 |
+
"loss": 1.8835,
|
1726 |
+
"step": 5060
|
1727 |
+
},
|
1728 |
+
{
|
1729 |
+
"epoch": 1.68,
|
1730 |
+
"learning_rate": 0.0,
|
1731 |
+
"loss": 1.741,
|
1732 |
+
"step": 5080
|
1733 |
+
},
|
1734 |
+
{
|
1735 |
+
"epoch": 1.69,
|
1736 |
+
"learning_rate": 0.0,
|
1737 |
+
"loss": 1.7543,
|
1738 |
+
"step": 5100
|
1739 |
+
},
|
1740 |
+
{
|
1741 |
+
"epoch": 1.7,
|
1742 |
+
"learning_rate": 0.0,
|
1743 |
+
"loss": 1.7913,
|
1744 |
+
"step": 5120
|
1745 |
+
},
|
1746 |
+
{
|
1747 |
+
"epoch": 1.7,
|
1748 |
+
"learning_rate": 0.0,
|
1749 |
+
"loss": 1.829,
|
1750 |
+
"step": 5140
|
1751 |
+
},
|
1752 |
+
{
|
1753 |
+
"epoch": 1.71,
|
1754 |
+
"learning_rate": 0.0,
|
1755 |
+
"loss": 1.7915,
|
1756 |
+
"step": 5160
|
1757 |
+
},
|
1758 |
+
{
|
1759 |
+
"epoch": 1.72,
|
1760 |
+
"learning_rate": 0.0,
|
1761 |
+
"loss": 1.767,
|
1762 |
+
"step": 5180
|
1763 |
+
},
|
1764 |
+
{
|
1765 |
+
"epoch": 1.72,
|
1766 |
+
"learning_rate": 0.0,
|
1767 |
+
"loss": 1.7809,
|
1768 |
+
"step": 5200
|
1769 |
+
},
|
1770 |
+
{
|
1771 |
+
"epoch": 1.72,
|
1772 |
+
"eval_loss": 1.970503568649292,
|
1773 |
+
"eval_runtime": 11951.295,
|
1774 |
+
"eval_samples_per_second": 1.119,
|
1775 |
+
"eval_steps_per_second": 0.14,
|
1776 |
+
"step": 5200
|
1777 |
+
},
|
1778 |
+
{
|
1779 |
+
"epoch": 1.73,
|
1780 |
+
"learning_rate": 0.0,
|
1781 |
+
"loss": 1.8715,
|
1782 |
+
"step": 5220
|
1783 |
+
},
|
1784 |
+
{
|
1785 |
+
"epoch": 1.74,
|
1786 |
+
"learning_rate": 0.0,
|
1787 |
+
"loss": 1.8517,
|
1788 |
+
"step": 5240
|
1789 |
+
},
|
1790 |
+
{
|
1791 |
+
"epoch": 1.74,
|
1792 |
+
"learning_rate": 0.0,
|
1793 |
+
"loss": 1.8255,
|
1794 |
+
"step": 5260
|
1795 |
+
},
|
1796 |
+
{
|
1797 |
+
"epoch": 1.75,
|
1798 |
+
"learning_rate": 0.0,
|
1799 |
+
"loss": 1.7973,
|
1800 |
+
"step": 5280
|
1801 |
+
},
|
1802 |
+
{
|
1803 |
+
"epoch": 1.76,
|
1804 |
+
"learning_rate": 0.0,
|
1805 |
+
"loss": 1.7912,
|
1806 |
+
"step": 5300
|
1807 |
+
},
|
1808 |
+
{
|
1809 |
+
"epoch": 1.76,
|
1810 |
+
"learning_rate": 0.0,
|
1811 |
+
"loss": 1.8331,
|
1812 |
+
"step": 5320
|
1813 |
+
},
|
1814 |
+
{
|
1815 |
+
"epoch": 1.77,
|
1816 |
+
"learning_rate": 0.0,
|
1817 |
+
"loss": 1.835,
|
1818 |
+
"step": 5340
|
1819 |
+
},
|
1820 |
+
{
|
1821 |
+
"epoch": 1.78,
|
1822 |
+
"learning_rate": 0.0,
|
1823 |
+
"loss": 1.8051,
|
1824 |
+
"step": 5360
|
1825 |
+
},
|
1826 |
+
{
|
1827 |
+
"epoch": 1.78,
|
1828 |
+
"learning_rate": 0.0,
|
1829 |
+
"loss": 1.8007,
|
1830 |
+
"step": 5380
|
1831 |
+
},
|
1832 |
+
{
|
1833 |
+
"epoch": 1.79,
|
1834 |
+
"learning_rate": 0.0,
|
1835 |
+
"loss": 1.8069,
|
1836 |
+
"step": 5400
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 1.79,
|
1840 |
+
"eval_loss": 1.970503568649292,
|
1841 |
+
"eval_runtime": 11948.5086,
|
1842 |
+
"eval_samples_per_second": 1.119,
|
1843 |
+
"eval_steps_per_second": 0.14,
|
1844 |
+
"step": 5400
|
1845 |
+
},
|
1846 |
+
{
|
1847 |
+
"epoch": 1.8,
|
1848 |
+
"learning_rate": 0.0,
|
1849 |
+
"loss": 1.784,
|
1850 |
+
"step": 5420
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 1.8,
|
1854 |
+
"learning_rate": 0.0,
|
1855 |
+
"loss": 1.7557,
|
1856 |
+
"step": 5440
|
1857 |
+
},
|
1858 |
+
{
|
1859 |
+
"epoch": 1.81,
|
1860 |
+
"learning_rate": 0.0,
|
1861 |
+
"loss": 1.8011,
|
1862 |
+
"step": 5460
|
1863 |
+
},
|
1864 |
+
{
|
1865 |
+
"epoch": 1.82,
|
1866 |
+
"learning_rate": 0.0,
|
1867 |
+
"loss": 1.8392,
|
1868 |
+
"step": 5480
|
1869 |
+
},
|
1870 |
+
{
|
1871 |
+
"epoch": 1.82,
|
1872 |
+
"learning_rate": 0.0,
|
1873 |
+
"loss": 1.7507,
|
1874 |
+
"step": 5500
|
1875 |
+
},
|
1876 |
+
{
|
1877 |
+
"epoch": 1.83,
|
1878 |
+
"learning_rate": 0.0,
|
1879 |
+
"loss": 1.8043,
|
1880 |
+
"step": 5520
|
1881 |
+
},
|
1882 |
+
{
|
1883 |
+
"epoch": 1.84,
|
1884 |
+
"learning_rate": 0.0,
|
1885 |
+
"loss": 1.7804,
|
1886 |
+
"step": 5540
|
1887 |
+
},
|
1888 |
+
{
|
1889 |
+
"epoch": 1.84,
|
1890 |
+
"learning_rate": 0.0,
|
1891 |
+
"loss": 1.811,
|
1892 |
+
"step": 5560
|
1893 |
+
},
|
1894 |
+
{
|
1895 |
+
"epoch": 1.85,
|
1896 |
+
"learning_rate": 0.0,
|
1897 |
+
"loss": 1.806,
|
1898 |
+
"step": 5580
|
1899 |
+
},
|
1900 |
+
{
|
1901 |
+
"epoch": 1.86,
|
1902 |
+
"learning_rate": 0.0,
|
1903 |
+
"loss": 1.8103,
|
1904 |
+
"step": 5600
|
1905 |
+
},
|
1906 |
+
{
|
1907 |
+
"epoch": 1.86,
|
1908 |
+
"eval_loss": 1.970503568649292,
|
1909 |
+
"eval_runtime": 11953.5582,
|
1910 |
+
"eval_samples_per_second": 1.118,
|
1911 |
+
"eval_steps_per_second": 0.14,
|
1912 |
+
"step": 5600
|
1913 |
+
},
|
1914 |
+
{
|
1915 |
+
"epoch": 1.86,
|
1916 |
+
"learning_rate": 0.0,
|
1917 |
+
"loss": 1.7947,
|
1918 |
+
"step": 5620
|
1919 |
+
},
|
1920 |
+
{
|
1921 |
+
"epoch": 1.87,
|
1922 |
+
"learning_rate": 0.0,
|
1923 |
+
"loss": 1.8062,
|
1924 |
+
"step": 5640
|
1925 |
+
},
|
1926 |
+
{
|
1927 |
+
"epoch": 1.88,
|
1928 |
+
"learning_rate": 0.0,
|
1929 |
+
"loss": 1.7997,
|
1930 |
+
"step": 5660
|
1931 |
+
},
|
1932 |
+
{
|
1933 |
+
"epoch": 1.88,
|
1934 |
+
"learning_rate": 0.0,
|
1935 |
+
"loss": 1.843,
|
1936 |
+
"step": 5680
|
1937 |
+
},
|
1938 |
+
{
|
1939 |
+
"epoch": 1.89,
|
1940 |
+
"learning_rate": 0.0,
|
1941 |
+
"loss": 1.848,
|
1942 |
+
"step": 5700
|
1943 |
+
},
|
1944 |
+
{
|
1945 |
+
"epoch": 1.9,
|
1946 |
+
"learning_rate": 0.0,
|
1947 |
+
"loss": 1.8393,
|
1948 |
+
"step": 5720
|
1949 |
+
},
|
1950 |
+
{
|
1951 |
+
"epoch": 1.9,
|
1952 |
+
"learning_rate": 0.0,
|
1953 |
+
"loss": 1.8187,
|
1954 |
+
"step": 5740
|
1955 |
+
},
|
1956 |
+
{
|
1957 |
+
"epoch": 1.91,
|
1958 |
+
"learning_rate": 0.0,
|
1959 |
+
"loss": 1.802,
|
1960 |
+
"step": 5760
|
1961 |
+
},
|
1962 |
+
{
|
1963 |
+
"epoch": 1.92,
|
1964 |
+
"learning_rate": 0.0,
|
1965 |
+
"loss": 1.7646,
|
1966 |
+
"step": 5780
|
1967 |
+
},
|
1968 |
+
{
|
1969 |
+
"epoch": 1.92,
|
1970 |
+
"learning_rate": 0.0,
|
1971 |
+
"loss": 1.7993,
|
1972 |
+
"step": 5800
|
1973 |
+
},
|
1974 |
+
{
|
1975 |
+
"epoch": 1.92,
|
1976 |
+
"eval_loss": 1.970503568649292,
|
1977 |
+
"eval_runtime": 11953.0956,
|
1978 |
+
"eval_samples_per_second": 1.118,
|
1979 |
+
"eval_steps_per_second": 0.14,
|
1980 |
+
"step": 5800
|
1981 |
+
},
|
1982 |
+
{
|
1983 |
+
"epoch": 1.93,
|
1984 |
+
"learning_rate": 0.0,
|
1985 |
+
"loss": 1.7962,
|
1986 |
+
"step": 5820
|
1987 |
+
},
|
1988 |
+
{
|
1989 |
+
"epoch": 1.94,
|
1990 |
+
"learning_rate": 0.0,
|
1991 |
+
"loss": 1.849,
|
1992 |
+
"step": 5840
|
1993 |
+
},
|
1994 |
+
{
|
1995 |
+
"epoch": 1.94,
|
1996 |
+
"learning_rate": 0.0,
|
1997 |
+
"loss": 1.8254,
|
1998 |
+
"step": 5860
|
1999 |
+
},
|
2000 |
+
{
|
2001 |
+
"epoch": 1.95,
|
2002 |
+
"learning_rate": 0.0,
|
2003 |
+
"loss": 1.8583,
|
2004 |
+
"step": 5880
|
2005 |
+
},
|
2006 |
+
{
|
2007 |
+
"epoch": 1.96,
|
2008 |
+
"learning_rate": 0.0,
|
2009 |
+
"loss": 1.8398,
|
2010 |
+
"step": 5900
|
2011 |
+
},
|
2012 |
+
{
|
2013 |
+
"epoch": 1.96,
|
2014 |
+
"learning_rate": 0.0,
|
2015 |
+
"loss": 1.7306,
|
2016 |
+
"step": 5920
|
2017 |
+
},
|
2018 |
+
{
|
2019 |
+
"epoch": 1.97,
|
2020 |
+
"learning_rate": 0.0,
|
2021 |
+
"loss": 1.81,
|
2022 |
+
"step": 5940
|
2023 |
+
},
|
2024 |
+
{
|
2025 |
+
"epoch": 1.98,
|
2026 |
+
"learning_rate": 0.0,
|
2027 |
+
"loss": 1.7762,
|
2028 |
+
"step": 5960
|
2029 |
+
},
|
2030 |
+
{
|
2031 |
+
"epoch": 1.98,
|
2032 |
+
"learning_rate": 0.0,
|
2033 |
+
"loss": 1.8549,
|
2034 |
+
"step": 5980
|
2035 |
+
},
|
2036 |
+
{
|
2037 |
+
"epoch": 1.99,
|
2038 |
+
"learning_rate": 0.0,
|
2039 |
+
"loss": 1.84,
|
2040 |
+
"step": 6000
|
2041 |
+
},
|
2042 |
+
{
|
2043 |
+
"epoch": 1.99,
|
2044 |
+
"eval_loss": 1.970503568649292,
|
2045 |
+
"eval_runtime": 11950.3766,
|
2046 |
+
"eval_samples_per_second": 1.119,
|
2047 |
+
"eval_steps_per_second": 0.14,
|
2048 |
+
"step": 6000
|
2049 |
+
},
|
2050 |
+
{
|
2051 |
+
"epoch": 2.0,
|
2052 |
+
"learning_rate": 0.0,
|
2053 |
+
"loss": 1.7897,
|
2054 |
+
"step": 6020
|
2055 |
+
},
|
2056 |
+
{
|
2057 |
+
"epoch": 2.0,
|
2058 |
+
"learning_rate": 0.0,
|
2059 |
+
"loss": 1.7844,
|
2060 |
+
"step": 6040
|
2061 |
+
},
|
2062 |
+
{
|
2063 |
+
"epoch": 2.01,
|
2064 |
+
"learning_rate": 0.0,
|
2065 |
+
"loss": 1.8453,
|
2066 |
+
"step": 6060
|
2067 |
+
},
|
2068 |
+
{
|
2069 |
+
"epoch": 2.02,
|
2070 |
+
"learning_rate": 0.0,
|
2071 |
+
"loss": 1.8268,
|
2072 |
+
"step": 6080
|
2073 |
+
},
|
2074 |
+
{
|
2075 |
+
"epoch": 2.02,
|
2076 |
+
"learning_rate": 0.0,
|
2077 |
+
"loss": 1.7932,
|
2078 |
+
"step": 6100
|
2079 |
+
},
|
2080 |
+
{
|
2081 |
+
"epoch": 2.03,
|
2082 |
+
"learning_rate": 0.0,
|
2083 |
+
"loss": 1.8005,
|
2084 |
+
"step": 6120
|
2085 |
+
},
|
2086 |
+
{
|
2087 |
+
"epoch": 2.04,
|
2088 |
+
"learning_rate": 0.0,
|
2089 |
+
"loss": 1.773,
|
2090 |
+
"step": 6140
|
2091 |
+
},
|
2092 |
+
{
|
2093 |
+
"epoch": 2.04,
|
2094 |
+
"learning_rate": 0.0,
|
2095 |
+
"loss": 1.8029,
|
2096 |
+
"step": 6160
|
2097 |
+
},
|
2098 |
+
{
|
2099 |
+
"epoch": 2.05,
|
2100 |
+
"learning_rate": 0.0,
|
2101 |
+
"loss": 1.8283,
|
2102 |
+
"step": 6180
|
2103 |
+
},
|
2104 |
+
{
|
2105 |
+
"epoch": 2.06,
|
2106 |
+
"learning_rate": 0.0,
|
2107 |
+
"loss": 1.8167,
|
2108 |
+
"step": 6200
|
2109 |
+
},
|
2110 |
+
{
|
2111 |
+
"epoch": 2.06,
|
2112 |
+
"eval_loss": 1.970503568649292,
|
2113 |
+
"eval_runtime": 11950.9199,
|
2114 |
+
"eval_samples_per_second": 1.119,
|
2115 |
+
"eval_steps_per_second": 0.14,
|
2116 |
+
"step": 6200
|
2117 |
+
},
|
2118 |
+
{
|
2119 |
+
"epoch": 2.06,
|
2120 |
+
"learning_rate": 0.0,
|
2121 |
+
"loss": 1.8465,
|
2122 |
+
"step": 6220
|
2123 |
+
},
|
2124 |
+
{
|
2125 |
+
"epoch": 2.07,
|
2126 |
+
"learning_rate": 0.0,
|
2127 |
+
"loss": 1.8747,
|
2128 |
+
"step": 6240
|
2129 |
+
},
|
2130 |
+
{
|
2131 |
+
"epoch": 2.08,
|
2132 |
+
"learning_rate": 0.0,
|
2133 |
+
"loss": 1.8031,
|
2134 |
+
"step": 6260
|
2135 |
+
},
|
2136 |
+
{
|
2137 |
+
"epoch": 2.08,
|
2138 |
+
"learning_rate": 0.0,
|
2139 |
+
"loss": 1.8366,
|
2140 |
+
"step": 6280
|
2141 |
+
},
|
2142 |
+
{
|
2143 |
+
"epoch": 2.09,
|
2144 |
+
"learning_rate": 0.0,
|
2145 |
+
"loss": 1.7998,
|
2146 |
+
"step": 6300
|
2147 |
+
},
|
2148 |
+
{
|
2149 |
+
"epoch": 2.1,
|
2150 |
+
"learning_rate": 0.0,
|
2151 |
+
"loss": 1.8143,
|
2152 |
+
"step": 6320
|
2153 |
+
},
|
2154 |
+
{
|
2155 |
+
"epoch": 2.1,
|
2156 |
+
"learning_rate": 0.0,
|
2157 |
+
"loss": 1.8586,
|
2158 |
+
"step": 6340
|
2159 |
+
},
|
2160 |
+
{
|
2161 |
+
"epoch": 2.11,
|
2162 |
+
"learning_rate": 0.0,
|
2163 |
+
"loss": 1.836,
|
2164 |
+
"step": 6360
|
2165 |
+
},
|
2166 |
+
{
|
2167 |
+
"epoch": 2.12,
|
2168 |
+
"learning_rate": 0.0,
|
2169 |
+
"loss": 1.863,
|
2170 |
+
"step": 6380
|
2171 |
+
},
|
2172 |
+
{
|
2173 |
+
"epoch": 2.12,
|
2174 |
+
"learning_rate": 0.0,
|
2175 |
+
"loss": 1.7667,
|
2176 |
+
"step": 6400
|
2177 |
+
},
|
2178 |
+
{
|
2179 |
+
"epoch": 2.12,
|
2180 |
+
"eval_loss": 1.970503568649292,
|
2181 |
+
"eval_runtime": 11949.7143,
|
2182 |
+
"eval_samples_per_second": 1.119,
|
2183 |
+
"eval_steps_per_second": 0.14,
|
2184 |
+
"step": 6400
|
2185 |
+
},
|
2186 |
+
{
|
2187 |
+
"epoch": 2.13,
|
2188 |
+
"learning_rate": 0.0,
|
2189 |
+
"loss": 1.8187,
|
2190 |
+
"step": 6420
|
2191 |
+
},
|
2192 |
+
{
|
2193 |
+
"epoch": 2.14,
|
2194 |
+
"learning_rate": 0.0,
|
2195 |
+
"loss": 1.7883,
|
2196 |
+
"step": 6440
|
2197 |
+
},
|
2198 |
+
{
|
2199 |
+
"epoch": 2.14,
|
2200 |
+
"learning_rate": 0.0,
|
2201 |
+
"loss": 1.9299,
|
2202 |
+
"step": 6460
|
2203 |
+
},
|
2204 |
+
{
|
2205 |
+
"epoch": 2.15,
|
2206 |
+
"learning_rate": 0.0,
|
2207 |
+
"loss": 1.8286,
|
2208 |
+
"step": 6480
|
2209 |
+
},
|
2210 |
+
{
|
2211 |
+
"epoch": 2.16,
|
2212 |
+
"learning_rate": 0.0,
|
2213 |
+
"loss": 1.8181,
|
2214 |
+
"step": 6500
|
2215 |
+
},
|
2216 |
+
{
|
2217 |
+
"epoch": 2.16,
|
2218 |
+
"learning_rate": 0.0,
|
2219 |
+
"loss": 1.7856,
|
2220 |
+
"step": 6520
|
2221 |
+
},
|
2222 |
+
{
|
2223 |
+
"epoch": 2.17,
|
2224 |
+
"learning_rate": 0.0,
|
2225 |
+
"loss": 1.7192,
|
2226 |
+
"step": 6540
|
2227 |
+
},
|
2228 |
+
{
|
2229 |
+
"epoch": 2.18,
|
2230 |
+
"learning_rate": 0.0,
|
2231 |
+
"loss": 1.7715,
|
2232 |
+
"step": 6560
|
2233 |
+
},
|
2234 |
+
{
|
2235 |
+
"epoch": 2.18,
|
2236 |
+
"learning_rate": 0.0,
|
2237 |
+
"loss": 1.8359,
|
2238 |
+
"step": 6580
|
2239 |
+
},
|
2240 |
+
{
|
2241 |
+
"epoch": 2.19,
|
2242 |
+
"learning_rate": 0.0,
|
2243 |
+
"loss": 1.7989,
|
2244 |
+
"step": 6600
|
2245 |
+
},
|
2246 |
+
{
|
2247 |
+
"epoch": 2.19,
|
2248 |
+
"eval_loss": 1.970503568649292,
|
2249 |
+
"eval_runtime": 11949.8024,
|
2250 |
+
"eval_samples_per_second": 1.119,
|
2251 |
+
"eval_steps_per_second": 0.14,
|
2252 |
+
"step": 6600
|
2253 |
+
},
|
2254 |
+
{
|
2255 |
+
"epoch": 2.2,
|
2256 |
+
"learning_rate": 0.0,
|
2257 |
+
"loss": 1.7869,
|
2258 |
+
"step": 6620
|
2259 |
+
},
|
2260 |
+
{
|
2261 |
+
"epoch": 2.2,
|
2262 |
+
"learning_rate": 0.0,
|
2263 |
+
"loss": 1.7855,
|
2264 |
+
"step": 6640
|
2265 |
+
},
|
2266 |
+
{
|
2267 |
+
"epoch": 2.21,
|
2268 |
+
"learning_rate": 0.0,
|
2269 |
+
"loss": 1.8312,
|
2270 |
+
"step": 6660
|
2271 |
+
},
|
2272 |
+
{
|
2273 |
+
"epoch": 2.22,
|
2274 |
+
"learning_rate": 0.0,
|
2275 |
+
"loss": 1.7717,
|
2276 |
+
"step": 6680
|
2277 |
+
},
|
2278 |
+
{
|
2279 |
+
"epoch": 2.22,
|
2280 |
+
"learning_rate": 0.0,
|
2281 |
+
"loss": 1.7621,
|
2282 |
+
"step": 6700
|
2283 |
+
},
|
2284 |
+
{
|
2285 |
+
"epoch": 2.23,
|
2286 |
+
"learning_rate": 0.0,
|
2287 |
+
"loss": 1.8031,
|
2288 |
+
"step": 6720
|
2289 |
+
},
|
2290 |
+
{
|
2291 |
+
"epoch": 2.24,
|
2292 |
+
"learning_rate": 0.0,
|
2293 |
+
"loss": 1.8061,
|
2294 |
+
"step": 6740
|
2295 |
+
},
|
2296 |
+
{
|
2297 |
+
"epoch": 2.24,
|
2298 |
+
"learning_rate": 0.0,
|
2299 |
+
"loss": 1.8063,
|
2300 |
+
"step": 6760
|
2301 |
+
},
|
2302 |
+
{
|
2303 |
+
"epoch": 2.25,
|
2304 |
+
"learning_rate": 0.0,
|
2305 |
+
"loss": 1.7469,
|
2306 |
+
"step": 6780
|
2307 |
+
},
|
2308 |
+
{
|
2309 |
+
"epoch": 2.26,
|
2310 |
+
"learning_rate": 0.0,
|
2311 |
+
"loss": 1.7543,
|
2312 |
+
"step": 6800
|
2313 |
+
},
|
2314 |
+
{
|
2315 |
+
"epoch": 2.26,
|
2316 |
+
"eval_loss": 1.970503568649292,
|
2317 |
+
"eval_runtime": 11954.2042,
|
2318 |
+
"eval_samples_per_second": 1.118,
|
2319 |
+
"eval_steps_per_second": 0.14,
|
2320 |
+
"step": 6800
|
2321 |
+
},
|
2322 |
+
{
|
2323 |
+
"epoch": 2.26,
|
2324 |
+
"learning_rate": 0.0,
|
2325 |
+
"loss": 1.8659,
|
2326 |
+
"step": 6820
|
2327 |
+
},
|
2328 |
+
{
|
2329 |
+
"epoch": 2.27,
|
2330 |
+
"learning_rate": 0.0,
|
2331 |
+
"loss": 1.7935,
|
2332 |
+
"step": 6840
|
2333 |
+
},
|
2334 |
+
{
|
2335 |
+
"epoch": 2.28,
|
2336 |
+
"learning_rate": 0.0,
|
2337 |
+
"loss": 1.8247,
|
2338 |
+
"step": 6860
|
2339 |
+
},
|
2340 |
+
{
|
2341 |
+
"epoch": 2.28,
|
2342 |
+
"learning_rate": 0.0,
|
2343 |
+
"loss": 1.8339,
|
2344 |
+
"step": 6880
|
2345 |
+
},
|
2346 |
+
{
|
2347 |
+
"epoch": 2.29,
|
2348 |
+
"learning_rate": 0.0,
|
2349 |
+
"loss": 1.8419,
|
2350 |
+
"step": 6900
|
2351 |
+
},
|
2352 |
+
{
|
2353 |
+
"epoch": 2.3,
|
2354 |
+
"learning_rate": 0.0,
|
2355 |
+
"loss": 1.7978,
|
2356 |
+
"step": 6920
|
2357 |
+
},
|
2358 |
+
{
|
2359 |
+
"epoch": 2.3,
|
2360 |
+
"learning_rate": 0.0,
|
2361 |
+
"loss": 1.8048,
|
2362 |
+
"step": 6940
|
2363 |
+
},
|
2364 |
+
{
|
2365 |
+
"epoch": 2.31,
|
2366 |
+
"learning_rate": 0.0,
|
2367 |
+
"loss": 1.8523,
|
2368 |
+
"step": 6960
|
2369 |
+
},
|
2370 |
+
{
|
2371 |
+
"epoch": 2.32,
|
2372 |
+
"learning_rate": 0.0,
|
2373 |
+
"loss": 1.7875,
|
2374 |
+
"step": 6980
|
2375 |
+
},
|
2376 |
+
{
|
2377 |
+
"epoch": 2.32,
|
2378 |
+
"learning_rate": 0.0,
|
2379 |
+
"loss": 1.8262,
|
2380 |
+
"step": 7000
|
2381 |
+
},
|
2382 |
+
{
|
2383 |
+
"epoch": 2.32,
|
2384 |
+
"eval_loss": 1.970503568649292,
|
2385 |
+
"eval_runtime": 11967.8262,
|
2386 |
+
"eval_samples_per_second": 1.117,
|
2387 |
+
"eval_steps_per_second": 0.14,
|
2388 |
+
"step": 7000
|
2389 |
+
},
|
2390 |
+
{
|
2391 |
+
"epoch": 2.33,
|
2392 |
+
"learning_rate": 0.0,
|
2393 |
+
"loss": 1.8661,
|
2394 |
+
"step": 7020
|
2395 |
+
},
|
2396 |
+
{
|
2397 |
+
"epoch": 2.33,
|
2398 |
+
"learning_rate": 0.0,
|
2399 |
+
"loss": 1.8136,
|
2400 |
+
"step": 7040
|
2401 |
+
},
|
2402 |
+
{
|
2403 |
+
"epoch": 2.34,
|
2404 |
+
"learning_rate": 0.0,
|
2405 |
+
"loss": 1.8066,
|
2406 |
+
"step": 7060
|
2407 |
+
},
|
2408 |
+
{
|
2409 |
+
"epoch": 2.35,
|
2410 |
+
"learning_rate": 0.0,
|
2411 |
+
"loss": 1.8355,
|
2412 |
+
"step": 7080
|
2413 |
+
},
|
2414 |
+
{
|
2415 |
+
"epoch": 2.35,
|
2416 |
+
"learning_rate": 0.0,
|
2417 |
+
"loss": 1.7598,
|
2418 |
+
"step": 7100
|
2419 |
+
},
|
2420 |
+
{
|
2421 |
+
"epoch": 2.36,
|
2422 |
+
"learning_rate": 0.0,
|
2423 |
+
"loss": 1.8384,
|
2424 |
+
"step": 7120
|
2425 |
+
},
|
2426 |
+
{
|
2427 |
+
"epoch": 2.37,
|
2428 |
+
"learning_rate": 0.0,
|
2429 |
+
"loss": 1.7768,
|
2430 |
+
"step": 7140
|
2431 |
+
},
|
2432 |
+
{
|
2433 |
+
"epoch": 2.37,
|
2434 |
+
"learning_rate": 0.0,
|
2435 |
+
"loss": 1.8371,
|
2436 |
+
"step": 7160
|
2437 |
+
},
|
2438 |
+
{
|
2439 |
+
"epoch": 2.38,
|
2440 |
+
"learning_rate": 0.0,
|
2441 |
+
"loss": 1.7989,
|
2442 |
+
"step": 7180
|
2443 |
+
},
|
2444 |
+
{
|
2445 |
+
"epoch": 2.39,
|
2446 |
+
"learning_rate": 0.0,
|
2447 |
+
"loss": 1.8204,
|
2448 |
+
"step": 7200
|
2449 |
+
},
|
2450 |
+
{
|
2451 |
+
"epoch": 2.39,
|
2452 |
+
"eval_loss": 1.970503568649292,
|
2453 |
+
"eval_runtime": 11959.9088,
|
2454 |
+
"eval_samples_per_second": 1.118,
|
2455 |
+
"eval_steps_per_second": 0.14,
|
2456 |
+
"step": 7200
|
2457 |
+
},
|
2458 |
+
{
|
2459 |
+
"epoch": 2.39,
|
2460 |
+
"learning_rate": 0.0,
|
2461 |
+
"loss": 1.8632,
|
2462 |
+
"step": 7220
|
2463 |
+
},
|
2464 |
+
{
|
2465 |
+
"epoch": 2.4,
|
2466 |
+
"learning_rate": 0.0,
|
2467 |
+
"loss": 1.838,
|
2468 |
+
"step": 7240
|
2469 |
+
},
|
2470 |
+
{
|
2471 |
+
"epoch": 2.41,
|
2472 |
+
"learning_rate": 0.0,
|
2473 |
+
"loss": 1.8492,
|
2474 |
+
"step": 7260
|
2475 |
+
},
|
2476 |
+
{
|
2477 |
+
"epoch": 2.41,
|
2478 |
+
"learning_rate": 0.0,
|
2479 |
+
"loss": 1.8213,
|
2480 |
+
"step": 7280
|
2481 |
+
},
|
2482 |
+
{
|
2483 |
+
"epoch": 2.42,
|
2484 |
+
"learning_rate": 0.0,
|
2485 |
+
"loss": 1.7367,
|
2486 |
+
"step": 7300
|
2487 |
+
},
|
2488 |
+
{
|
2489 |
+
"epoch": 2.43,
|
2490 |
+
"learning_rate": 0.0,
|
2491 |
+
"loss": 1.9046,
|
2492 |
+
"step": 7320
|
2493 |
+
},
|
2494 |
+
{
|
2495 |
+
"epoch": 2.43,
|
2496 |
+
"learning_rate": 0.0,
|
2497 |
+
"loss": 1.7799,
|
2498 |
+
"step": 7340
|
2499 |
+
},
|
2500 |
+
{
|
2501 |
+
"epoch": 2.44,
|
2502 |
+
"learning_rate": 0.0,
|
2503 |
+
"loss": 1.793,
|
2504 |
+
"step": 7360
|
2505 |
+
},
|
2506 |
+
{
|
2507 |
+
"epoch": 2.45,
|
2508 |
+
"learning_rate": 0.0,
|
2509 |
+
"loss": 1.7864,
|
2510 |
+
"step": 7380
|
2511 |
+
},
|
2512 |
+
{
|
2513 |
+
"epoch": 2.45,
|
2514 |
+
"learning_rate": 0.0,
|
2515 |
+
"loss": 1.8071,
|
2516 |
+
"step": 7400
|
2517 |
+
},
|
2518 |
+
{
|
2519 |
+
"epoch": 2.45,
|
2520 |
+
"eval_loss": 1.970503568649292,
|
2521 |
+
"eval_runtime": 11949.9374,
|
2522 |
+
"eval_samples_per_second": 1.119,
|
2523 |
+
"eval_steps_per_second": 0.14,
|
2524 |
+
"step": 7400
|
2525 |
+
},
|
2526 |
+
{
|
2527 |
+
"epoch": 2.46,
|
2528 |
+
"learning_rate": 0.0,
|
2529 |
+
"loss": 1.7708,
|
2530 |
+
"step": 7420
|
2531 |
+
},
|
2532 |
+
{
|
2533 |
+
"epoch": 2.47,
|
2534 |
+
"learning_rate": 0.0,
|
2535 |
+
"loss": 1.8234,
|
2536 |
+
"step": 7440
|
2537 |
+
},
|
2538 |
+
{
|
2539 |
+
"epoch": 2.47,
|
2540 |
+
"learning_rate": 0.0,
|
2541 |
+
"loss": 1.8214,
|
2542 |
+
"step": 7460
|
2543 |
+
},
|
2544 |
+
{
|
2545 |
+
"epoch": 2.48,
|
2546 |
+
"learning_rate": 0.0,
|
2547 |
+
"loss": 1.8468,
|
2548 |
+
"step": 7480
|
2549 |
+
},
|
2550 |
+
{
|
2551 |
+
"epoch": 2.49,
|
2552 |
+
"learning_rate": 0.0,
|
2553 |
+
"loss": 1.784,
|
2554 |
+
"step": 7500
|
2555 |
+
},
|
2556 |
+
{
|
2557 |
+
"epoch": 2.49,
|
2558 |
+
"learning_rate": 0.0,
|
2559 |
+
"loss": 1.8571,
|
2560 |
+
"step": 7520
|
2561 |
+
},
|
2562 |
+
{
|
2563 |
+
"epoch": 2.5,
|
2564 |
+
"learning_rate": 0.0,
|
2565 |
+
"loss": 1.7871,
|
2566 |
+
"step": 7540
|
2567 |
+
},
|
2568 |
+
{
|
2569 |
+
"epoch": 2.51,
|
2570 |
+
"learning_rate": 0.0,
|
2571 |
+
"loss": 1.7819,
|
2572 |
+
"step": 7560
|
2573 |
+
},
|
2574 |
+
{
|
2575 |
+
"epoch": 2.51,
|
2576 |
+
"learning_rate": 0.0,
|
2577 |
+
"loss": 1.8035,
|
2578 |
+
"step": 7580
|
2579 |
+
},
|
2580 |
+
{
|
2581 |
+
"epoch": 2.52,
|
2582 |
+
"learning_rate": 0.0,
|
2583 |
+
"loss": 1.801,
|
2584 |
+
"step": 7600
|
2585 |
+
},
|
2586 |
+
{
|
2587 |
+
"epoch": 2.52,
|
2588 |
+
"eval_loss": 1.970503568649292,
|
2589 |
+
"eval_runtime": 11949.3183,
|
2590 |
+
"eval_samples_per_second": 1.119,
|
2591 |
+
"eval_steps_per_second": 0.14,
|
2592 |
+
"step": 7600
|
2593 |
+
},
|
2594 |
+
{
|
2595 |
+
"epoch": 2.53,
|
2596 |
+
"learning_rate": 0.0,
|
2597 |
+
"loss": 1.8302,
|
2598 |
+
"step": 7620
|
2599 |
+
},
|
2600 |
+
{
|
2601 |
+
"epoch": 2.53,
|
2602 |
+
"learning_rate": 0.0,
|
2603 |
+
"loss": 1.8208,
|
2604 |
+
"step": 7640
|
2605 |
+
},
|
2606 |
+
{
|
2607 |
+
"epoch": 2.54,
|
2608 |
+
"learning_rate": 0.0,
|
2609 |
+
"loss": 1.917,
|
2610 |
+
"step": 7660
|
2611 |
+
},
|
2612 |
+
{
|
2613 |
+
"epoch": 2.55,
|
2614 |
+
"learning_rate": 0.0,
|
2615 |
+
"loss": 1.8184,
|
2616 |
+
"step": 7680
|
2617 |
+
},
|
2618 |
+
{
|
2619 |
+
"epoch": 2.55,
|
2620 |
+
"learning_rate": 0.0,
|
2621 |
+
"loss": 1.7462,
|
2622 |
+
"step": 7700
|
2623 |
+
},
|
2624 |
+
{
|
2625 |
+
"epoch": 2.56,
|
2626 |
+
"learning_rate": 0.0,
|
2627 |
+
"loss": 1.7699,
|
2628 |
+
"step": 7720
|
2629 |
+
},
|
2630 |
+
{
|
2631 |
+
"epoch": 2.57,
|
2632 |
+
"learning_rate": 0.0,
|
2633 |
+
"loss": 1.8377,
|
2634 |
+
"step": 7740
|
2635 |
+
},
|
2636 |
+
{
|
2637 |
+
"epoch": 2.57,
|
2638 |
+
"learning_rate": 0.0,
|
2639 |
+
"loss": 1.7713,
|
2640 |
+
"step": 7760
|
2641 |
+
},
|
2642 |
+
{
|
2643 |
+
"epoch": 2.58,
|
2644 |
+
"learning_rate": 0.0,
|
2645 |
+
"loss": 1.798,
|
2646 |
+
"step": 7780
|
2647 |
+
},
|
2648 |
+
{
|
2649 |
+
"epoch": 2.59,
|
2650 |
+
"learning_rate": 0.0,
|
2651 |
+
"loss": 1.8425,
|
2652 |
+
"step": 7800
|
2653 |
+
},
|
2654 |
+
{
|
2655 |
+
"epoch": 2.59,
|
2656 |
+
"eval_loss": 1.970503568649292,
|
2657 |
+
"eval_runtime": 11947.7265,
|
2658 |
+
"eval_samples_per_second": 1.119,
|
2659 |
+
"eval_steps_per_second": 0.14,
|
2660 |
+
"step": 7800
|
2661 |
+
},
|
2662 |
+
{
|
2663 |
+
"epoch": 2.59,
|
2664 |
+
"learning_rate": 0.0,
|
2665 |
+
"loss": 1.7816,
|
2666 |
+
"step": 7820
|
2667 |
+
},
|
2668 |
+
{
|
2669 |
+
"epoch": 2.6,
|
2670 |
+
"learning_rate": 0.0,
|
2671 |
+
"loss": 1.8193,
|
2672 |
+
"step": 7840
|
2673 |
+
},
|
2674 |
+
{
|
2675 |
+
"epoch": 2.61,
|
2676 |
+
"learning_rate": 0.0,
|
2677 |
+
"loss": 1.8044,
|
2678 |
+
"step": 7860
|
2679 |
+
},
|
2680 |
+
{
|
2681 |
+
"epoch": 2.61,
|
2682 |
+
"learning_rate": 0.0,
|
2683 |
+
"loss": 1.7937,
|
2684 |
+
"step": 7880
|
2685 |
+
},
|
2686 |
+
{
|
2687 |
+
"epoch": 2.62,
|
2688 |
+
"learning_rate": 0.0,
|
2689 |
+
"loss": 1.8224,
|
2690 |
+
"step": 7900
|
2691 |
+
},
|
2692 |
+
{
|
2693 |
+
"epoch": 2.63,
|
2694 |
+
"learning_rate": 0.0,
|
2695 |
+
"loss": 1.7937,
|
2696 |
+
"step": 7920
|
2697 |
+
},
|
2698 |
+
{
|
2699 |
+
"epoch": 2.63,
|
2700 |
+
"learning_rate": 0.0,
|
2701 |
+
"loss": 1.7988,
|
2702 |
+
"step": 7940
|
2703 |
+
},
|
2704 |
+
{
|
2705 |
+
"epoch": 2.64,
|
2706 |
+
"learning_rate": 0.0,
|
2707 |
+
"loss": 1.8622,
|
2708 |
+
"step": 7960
|
2709 |
+
},
|
2710 |
+
{
|
2711 |
+
"epoch": 2.65,
|
2712 |
+
"learning_rate": 0.0,
|
2713 |
+
"loss": 1.8407,
|
2714 |
+
"step": 7980
|
2715 |
+
},
|
2716 |
+
{
|
2717 |
+
"epoch": 2.65,
|
2718 |
+
"learning_rate": 0.0,
|
2719 |
+
"loss": 1.7578,
|
2720 |
+
"step": 8000
|
2721 |
+
},
|
2722 |
+
{
|
2723 |
+
"epoch": 2.65,
|
2724 |
+
"eval_loss": 1.970503568649292,
|
2725 |
+
"eval_runtime": 11948.4274,
|
2726 |
+
"eval_samples_per_second": 1.119,
|
2727 |
+
"eval_steps_per_second": 0.14,
|
2728 |
+
"step": 8000
|
2729 |
+
},
|
2730 |
+
{
|
2731 |
+
"epoch": 2.66,
|
2732 |
+
"learning_rate": 0.0,
|
2733 |
+
"loss": 1.789,
|
2734 |
+
"step": 8020
|
2735 |
+
},
|
2736 |
+
{
|
2737 |
+
"epoch": 2.67,
|
2738 |
+
"learning_rate": 0.0,
|
2739 |
+
"loss": 1.7939,
|
2740 |
+
"step": 8040
|
2741 |
+
},
|
2742 |
+
{
|
2743 |
+
"epoch": 2.67,
|
2744 |
+
"learning_rate": 0.0,
|
2745 |
+
"loss": 1.7694,
|
2746 |
+
"step": 8060
|
2747 |
+
},
|
2748 |
+
{
|
2749 |
+
"epoch": 2.68,
|
2750 |
+
"learning_rate": 0.0,
|
2751 |
+
"loss": 1.7917,
|
2752 |
+
"step": 8080
|
2753 |
+
},
|
2754 |
+
{
|
2755 |
+
"epoch": 2.69,
|
2756 |
+
"learning_rate": 0.0,
|
2757 |
+
"loss": 1.7737,
|
2758 |
+
"step": 8100
|
2759 |
+
},
|
2760 |
+
{
|
2761 |
+
"epoch": 2.69,
|
2762 |
+
"learning_rate": 0.0,
|
2763 |
+
"loss": 1.8049,
|
2764 |
+
"step": 8120
|
2765 |
+
},
|
2766 |
+
{
|
2767 |
+
"epoch": 2.7,
|
2768 |
+
"learning_rate": 0.0,
|
2769 |
+
"loss": 1.8419,
|
2770 |
+
"step": 8140
|
2771 |
+
},
|
2772 |
+
{
|
2773 |
+
"epoch": 2.71,
|
2774 |
+
"learning_rate": 0.0,
|
2775 |
+
"loss": 1.7835,
|
2776 |
+
"step": 8160
|
2777 |
+
},
|
2778 |
+
{
|
2779 |
+
"epoch": 2.71,
|
2780 |
+
"learning_rate": 0.0,
|
2781 |
+
"loss": 1.7898,
|
2782 |
+
"step": 8180
|
2783 |
+
},
|
2784 |
+
{
|
2785 |
+
"epoch": 2.72,
|
2786 |
+
"learning_rate": 0.0,
|
2787 |
+
"loss": 1.7798,
|
2788 |
+
"step": 8200
|
2789 |
+
},
|
2790 |
+
{
|
2791 |
+
"epoch": 2.72,
|
2792 |
+
"eval_loss": 1.970503568649292,
|
2793 |
+
"eval_runtime": 11959.3563,
|
2794 |
+
"eval_samples_per_second": 1.118,
|
2795 |
+
"eval_steps_per_second": 0.14,
|
2796 |
+
"step": 8200
|
2797 |
+
},
|
2798 |
+
{
|
2799 |
+
"epoch": 2.73,
|
2800 |
+
"learning_rate": 0.0,
|
2801 |
+
"loss": 1.7904,
|
2802 |
+
"step": 8220
|
2803 |
+
},
|
2804 |
+
{
|
2805 |
+
"epoch": 2.73,
|
2806 |
+
"learning_rate": 0.0,
|
2807 |
+
"loss": 1.8054,
|
2808 |
+
"step": 8240
|
2809 |
+
},
|
2810 |
+
{
|
2811 |
+
"epoch": 2.74,
|
2812 |
+
"learning_rate": 0.0,
|
2813 |
+
"loss": 1.8316,
|
2814 |
+
"step": 8260
|
2815 |
+
},
|
2816 |
+
{
|
2817 |
+
"epoch": 2.75,
|
2818 |
+
"learning_rate": 0.0,
|
2819 |
+
"loss": 1.8059,
|
2820 |
+
"step": 8280
|
2821 |
+
},
|
2822 |
+
{
|
2823 |
+
"epoch": 2.75,
|
2824 |
+
"learning_rate": 0.0,
|
2825 |
+
"loss": 1.7908,
|
2826 |
+
"step": 8300
|
2827 |
+
},
|
2828 |
+
{
|
2829 |
+
"epoch": 2.76,
|
2830 |
+
"learning_rate": 0.0,
|
2831 |
+
"loss": 1.8092,
|
2832 |
+
"step": 8320
|
2833 |
+
},
|
2834 |
+
{
|
2835 |
+
"epoch": 2.77,
|
2836 |
+
"learning_rate": 0.0,
|
2837 |
+
"loss": 1.8954,
|
2838 |
+
"step": 8340
|
2839 |
+
},
|
2840 |
+
{
|
2841 |
+
"epoch": 2.77,
|
2842 |
+
"learning_rate": 0.0,
|
2843 |
+
"loss": 1.835,
|
2844 |
+
"step": 8360
|
2845 |
+
},
|
2846 |
+
{
|
2847 |
+
"epoch": 2.78,
|
2848 |
+
"learning_rate": 0.0,
|
2849 |
+
"loss": 1.8018,
|
2850 |
+
"step": 8380
|
2851 |
+
},
|
2852 |
+
{
|
2853 |
+
"epoch": 2.79,
|
2854 |
+
"learning_rate": 0.0,
|
2855 |
+
"loss": 1.8474,
|
2856 |
+
"step": 8400
|
2857 |
+
},
|
2858 |
+
{
|
2859 |
+
"epoch": 2.79,
|
2860 |
+
"eval_loss": 1.970503568649292,
|
2861 |
+
"eval_runtime": 11978.111,
|
2862 |
+
"eval_samples_per_second": 1.116,
|
2863 |
+
"eval_steps_per_second": 0.14,
|
2864 |
+
"step": 8400
|
2865 |
+
},
|
2866 |
+
{
|
2867 |
+
"epoch": 2.79,
|
2868 |
+
"learning_rate": 0.0,
|
2869 |
+
"loss": 1.7878,
|
2870 |
+
"step": 8420
|
2871 |
+
},
|
2872 |
+
{
|
2873 |
+
"epoch": 2.8,
|
2874 |
+
"learning_rate": 0.0,
|
2875 |
+
"loss": 1.7629,
|
2876 |
+
"step": 8440
|
2877 |
+
},
|
2878 |
+
{
|
2879 |
+
"epoch": 2.81,
|
2880 |
+
"learning_rate": 0.0,
|
2881 |
+
"loss": 1.8068,
|
2882 |
+
"step": 8460
|
2883 |
+
},
|
2884 |
+
{
|
2885 |
+
"epoch": 2.81,
|
2886 |
+
"learning_rate": 0.0,
|
2887 |
+
"loss": 1.7907,
|
2888 |
+
"step": 8480
|
2889 |
+
},
|
2890 |
+
{
|
2891 |
+
"epoch": 2.82,
|
2892 |
+
"learning_rate": 0.0,
|
2893 |
+
"loss": 1.7598,
|
2894 |
+
"step": 8500
|
2895 |
+
},
|
2896 |
+
{
|
2897 |
+
"epoch": 2.83,
|
2898 |
+
"learning_rate": 0.0,
|
2899 |
+
"loss": 1.7964,
|
2900 |
+
"step": 8520
|
2901 |
+
},
|
2902 |
+
{
|
2903 |
+
"epoch": 2.83,
|
2904 |
+
"learning_rate": 0.0,
|
2905 |
+
"loss": 1.7951,
|
2906 |
+
"step": 8540
|
2907 |
+
},
|
2908 |
+
{
|
2909 |
+
"epoch": 2.84,
|
2910 |
+
"learning_rate": 0.0,
|
2911 |
+
"loss": 1.7962,
|
2912 |
+
"step": 8560
|
2913 |
+
},
|
2914 |
+
{
|
2915 |
+
"epoch": 2.85,
|
2916 |
+
"learning_rate": 0.0,
|
2917 |
+
"loss": 1.7908,
|
2918 |
+
"step": 8580
|
2919 |
+
},
|
2920 |
+
{
|
2921 |
+
"epoch": 2.85,
|
2922 |
+
"learning_rate": 0.0,
|
2923 |
+
"loss": 1.7525,
|
2924 |
+
"step": 8600
|
2925 |
+
},
|
2926 |
+
{
|
2927 |
+
"epoch": 2.85,
|
2928 |
+
"eval_loss": 1.970503568649292,
|
2929 |
+
"eval_runtime": 11977.8941,
|
2930 |
+
"eval_samples_per_second": 1.116,
|
2931 |
+
"eval_steps_per_second": 0.14,
|
2932 |
+
"step": 8600
|
2933 |
+
},
|
2934 |
+
{
|
2935 |
+
"epoch": 2.86,
|
2936 |
+
"learning_rate": 0.0,
|
2937 |
+
"loss": 1.7268,
|
2938 |
+
"step": 8620
|
2939 |
+
},
|
2940 |
+
{
|
2941 |
+
"epoch": 2.87,
|
2942 |
+
"learning_rate": 0.0,
|
2943 |
+
"loss": 1.7716,
|
2944 |
+
"step": 8640
|
2945 |
+
},
|
2946 |
+
{
|
2947 |
+
"epoch": 2.87,
|
2948 |
+
"learning_rate": 0.0,
|
2949 |
+
"loss": 1.8214,
|
2950 |
+
"step": 8660
|
2951 |
+
},
|
2952 |
+
{
|
2953 |
+
"epoch": 2.88,
|
2954 |
+
"learning_rate": 0.0,
|
2955 |
+
"loss": 1.8116,
|
2956 |
+
"step": 8680
|
2957 |
+
},
|
2958 |
+
{
|
2959 |
+
"epoch": 2.89,
|
2960 |
+
"learning_rate": 0.0,
|
2961 |
+
"loss": 1.8204,
|
2962 |
+
"step": 8700
|
2963 |
+
},
|
2964 |
+
{
|
2965 |
+
"epoch": 2.89,
|
2966 |
+
"learning_rate": 0.0,
|
2967 |
+
"loss": 1.7878,
|
2968 |
+
"step": 8720
|
2969 |
+
},
|
2970 |
+
{
|
2971 |
+
"epoch": 2.9,
|
2972 |
+
"learning_rate": 0.0,
|
2973 |
+
"loss": 1.8828,
|
2974 |
+
"step": 8740
|
2975 |
+
},
|
2976 |
+
{
|
2977 |
+
"epoch": 2.91,
|
2978 |
+
"learning_rate": 0.0,
|
2979 |
+
"loss": 1.8015,
|
2980 |
+
"step": 8760
|
2981 |
+
},
|
2982 |
+
{
|
2983 |
+
"epoch": 2.91,
|
2984 |
+
"learning_rate": 0.0,
|
2985 |
+
"loss": 1.7989,
|
2986 |
+
"step": 8780
|
2987 |
+
},
|
2988 |
+
{
|
2989 |
+
"epoch": 2.92,
|
2990 |
+
"learning_rate": 0.0,
|
2991 |
+
"loss": 1.7467,
|
2992 |
+
"step": 8800
|
2993 |
+
},
|
2994 |
+
{
|
2995 |
+
"epoch": 2.92,
|
2996 |
+
"eval_loss": 1.970503568649292,
|
2997 |
+
"eval_runtime": 11960.8065,
|
2998 |
+
"eval_samples_per_second": 1.118,
|
2999 |
+
"eval_steps_per_second": 0.14,
|
3000 |
+
"step": 8800
|
3001 |
+
},
|
3002 |
+
{
|
3003 |
+
"epoch": 2.93,
|
3004 |
+
"learning_rate": 0.0,
|
3005 |
+
"loss": 1.7823,
|
3006 |
+
"step": 8820
|
3007 |
+
},
|
3008 |
+
{
|
3009 |
+
"epoch": 2.93,
|
3010 |
+
"learning_rate": 0.0,
|
3011 |
+
"loss": 1.8734,
|
3012 |
+
"step": 8840
|
3013 |
+
},
|
3014 |
+
{
|
3015 |
+
"epoch": 2.94,
|
3016 |
+
"learning_rate": 0.0,
|
3017 |
+
"loss": 1.8192,
|
3018 |
+
"step": 8860
|
3019 |
+
},
|
3020 |
+
{
|
3021 |
+
"epoch": 2.95,
|
3022 |
+
"learning_rate": 0.0,
|
3023 |
+
"loss": 1.8,
|
3024 |
+
"step": 8880
|
3025 |
+
},
|
3026 |
+
{
|
3027 |
+
"epoch": 2.95,
|
3028 |
+
"learning_rate": 0.0,
|
3029 |
+
"loss": 1.8057,
|
3030 |
+
"step": 8900
|
3031 |
+
},
|
3032 |
+
{
|
3033 |
+
"epoch": 2.96,
|
3034 |
+
"learning_rate": 0.0,
|
3035 |
+
"loss": 1.8007,
|
3036 |
+
"step": 8920
|
3037 |
+
},
|
3038 |
+
{
|
3039 |
+
"epoch": 2.97,
|
3040 |
+
"learning_rate": 0.0,
|
3041 |
+
"loss": 1.7826,
|
3042 |
+
"step": 8940
|
3043 |
+
},
|
3044 |
+
{
|
3045 |
+
"epoch": 2.97,
|
3046 |
+
"learning_rate": 0.0,
|
3047 |
+
"loss": 1.8612,
|
3048 |
+
"step": 8960
|
3049 |
+
},
|
3050 |
+
{
|
3051 |
+
"epoch": 2.98,
|
3052 |
+
"learning_rate": 0.0,
|
3053 |
+
"loss": 1.7843,
|
3054 |
+
"step": 8980
|
3055 |
+
},
|
3056 |
+
{
|
3057 |
+
"epoch": 2.99,
|
3058 |
+
"learning_rate": 0.0,
|
3059 |
+
"loss": 1.8619,
|
3060 |
+
"step": 9000
|
3061 |
+
},
|
3062 |
+
{
|
3063 |
+
"epoch": 2.99,
|
3064 |
+
"eval_loss": 1.970503568649292,
|
3065 |
+
"eval_runtime": 11953.5723,
|
3066 |
+
"eval_samples_per_second": 1.118,
|
3067 |
+
"eval_steps_per_second": 0.14,
|
3068 |
+
"step": 9000
|
3069 |
+
}
|
3070 |
+
],
|
3071 |
+
"max_steps": 9045,
|
3072 |
+
"num_train_epochs": 3,
|
3073 |
+
"total_flos": 5.6126739980068454e+17,
|
3074 |
+
"trial_name": null,
|
3075 |
+
"trial_params": null
|
3076 |
+
}
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail-results_6000_samples/checkpoint-9000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f5a78612ea8930d68eba4cb53d62254ccf547582e754aa049d169c3c11dd5fe4
|
3 |
+
size 4027
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail_6000_samples/adapter_config.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"base_model_name_or_path": "/domino/edv/afs-mrmc-data-store-rw/innovation/hf/RedPajama-INCITE-7B-Base",
|
3 |
+
"bias": "none",
|
4 |
+
"fan_in_fan_out": false,
|
5 |
+
"inference_mode": true,
|
6 |
+
"init_lora_weights": true,
|
7 |
+
"lora_alpha": 16,
|
8 |
+
"lora_dropout": 0.05,
|
9 |
+
"modules_to_save": null,
|
10 |
+
"peft_type": "LORA",
|
11 |
+
"r": 8,
|
12 |
+
"target_modules": [
|
13 |
+
"query_key_value"
|
14 |
+
],
|
15 |
+
"task_type": "CAUSAL_LM"
|
16 |
+
}
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail_6000_samples/adapter_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7004b69a5e5338ee5dbc682271079d0cf6750a64f3dd06f0dabf0a4c8129f41b
|
3 |
+
size 16800753
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail_6000_samples/special_tokens_map.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<|endoftext|>",
|
3 |
+
"eos_token": "<eos>",
|
4 |
+
"pad_token": "<|endoftext|>",
|
5 |
+
"unk_token": "<|endoftext|>"
|
6 |
+
}
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail_6000_samples/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
redpj7B-lora-cnn-dailymail_6000_samples/results/redpj7B-lora-cnn-dailymail_6000_samples/tokenizer_config.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_eos_token": true,
|
3 |
+
"add_prefix_space": false,
|
4 |
+
"bos_token": "<|endoftext|>",
|
5 |
+
"clean_up_tokenization_spaces": true,
|
6 |
+
"eos_token": "<|endoftext|>",
|
7 |
+
"model_max_length": 2048,
|
8 |
+
"tokenizer_class": "GPTNeoXTokenizer",
|
9 |
+
"unk_token": "<|endoftext|>"
|
10 |
+
}
|
redpj7B-lora-cnn-dailymail_6000_samples/results/stdout.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
redpj7B-lora-cnn-dailymail_6000_samples/script_fine_tuning.py
ADDED
@@ -0,0 +1,170 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
afs_path = '/domino/edv/afs-mrmc-data-store-rw/innovation/hf/'
|
2 |
+
|
3 |
+
import datasets
|
4 |
+
from datasets import load_dataset
|
5 |
+
import numpy as np
|
6 |
+
|
7 |
+
from peft import LoraConfig, get_peft_model, prepare_model_for_int8_training, TaskType, PeftModel
|
8 |
+
|
9 |
+
import transformers
|
10 |
+
import torch
|
11 |
+
print('transformers version: '+transformers.__version__)
|
12 |
+
#print('tensorflow version: '+tf.__version__)
|
13 |
+
print('torch version: '+torch.__version__)
|
14 |
+
|
15 |
+
|
16 |
+
device = 'cuda' if torch.cuda.is_available() else 'cpu'
|
17 |
+
|
18 |
+
|
19 |
+
model = '7B' #'7B' # Pick your poison
|
20 |
+
|
21 |
+
if model == '7B':
|
22 |
+
model_name = ("RedPajama-INCITE-7B-Base","RedPajama-INCITE-7B-Base")
|
23 |
+
run_name = 'redpj7B-lora-cnn-dailymail_fine_tune_test'
|
24 |
+
dataset = 'cnn_dailymail'
|
25 |
+
peft_name = './results/redpj7B-lora-cnn-dailymail_fine_tune_test'
|
26 |
+
output_dir = './results/redpj7B-lora-cnn-dailymail-results_fine_tune_test'
|
27 |
+
else: #3B
|
28 |
+
model_name = ("RedPajama-INCITE-Base-3B-v1","RedPajama-INCITE-Base-3B-v1")
|
29 |
+
run_name = 'redpj3B-lora-cnn-dailymail_fine_tune_test'
|
30 |
+
dataset = 'cnn_dailymail'
|
31 |
+
peft_name = './results/redpj3B-lora-cnn-dailymail_fine_tune_test'
|
32 |
+
output_dir = './results/redpj3B-lora-cnn-dailymail-results_fine_tune_test'
|
33 |
+
|
34 |
+
print(f"""model_name: {model_name[1]}, dataset: {dataset}, peft_name {peft_name}, run_name {run_name}, output_dir {output_dir}""")
|
35 |
+
|
36 |
+
|
37 |
+
from transformers import AutoTokenizer
|
38 |
+
|
39 |
+
print("Loading tokenizer for model: ", model_name[1])
|
40 |
+
tokenizer = AutoTokenizer.from_pretrained(afs_path+model_name[1],add_eos_token=True)
|
41 |
+
tokenizer.pad_token_id = 0
|
42 |
+
|
43 |
+
tokenizer.add_special_tokens({'eos_token':'<eos>'})
|
44 |
+
print('eos_token_id:',tokenizer.eos_token_id)
|
45 |
+
|
46 |
+
#CUTOFF_LEN = 256 # 256 accounts for about 96% of the data in the alpaca dataset
|
47 |
+
CUTOFF_LEN = 781 # 781 is the average token count for the articles according to https://huggingface.co/datasets/cnn_dailymail
|
48 |
+
|
49 |
+
|
50 |
+
def tokenize(prompt, tokenizer,add_eos_token=True):
|
51 |
+
result = tokenizer(
|
52 |
+
prompt+"<eos>", # add the end-of-stream token
|
53 |
+
truncation=True,
|
54 |
+
max_length=CUTOFF_LEN,
|
55 |
+
padding="max_length",
|
56 |
+
)
|
57 |
+
return {
|
58 |
+
"input_ids": result["input_ids"],
|
59 |
+
"attention_mask": result["attention_mask"],
|
60 |
+
}
|
61 |
+
|
62 |
+
|
63 |
+
|
64 |
+
data = datasets.load_from_disk('cnn_dailymail_dataset')
|
65 |
+
|
66 |
+
num_train_examples = len(data['train'])
|
67 |
+
|
68 |
+
# Define the percentage of data you want to keep
|
69 |
+
percentage_to_keep = 0.02 # Adjust this value to your desired percentage (0.02 is about 6k samples)
|
70 |
+
|
71 |
+
# Calculate the number of examples to keep
|
72 |
+
num_examples_to_keep = int(num_train_examples * percentage_to_keep)
|
73 |
+
|
74 |
+
# Reduce the 'train' split to the desired amount
|
75 |
+
train_data_reduced = data['train'].select(range(num_examples_to_keep))
|
76 |
+
|
77 |
+
#train_data_reduced.save_to_disk("./cnn_dailymail_dataset/train_data_reduced")
|
78 |
+
|
79 |
+
def generate_prompt(data_point):
|
80 |
+
# sorry about the formatting disaster gotta move fast
|
81 |
+
if data_point["article"]:
|
82 |
+
return f"""Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
|
83 |
+
|
84 |
+
### Instruction:
|
85 |
+
Summarize the text from the input.
|
86 |
+
|
87 |
+
### Input:
|
88 |
+
{data_point["article"]}
|
89 |
+
|
90 |
+
### Response:
|
91 |
+
{data_point["highlights"]}"""
|
92 |
+
else:
|
93 |
+
return f"""Below is an instruction that describes a task. Write a response that appropriately completes the request.
|
94 |
+
|
95 |
+
### Instruction:
|
96 |
+
Summarize the text.
|
97 |
+
|
98 |
+
### Response:
|
99 |
+
{data_point["highlights"]}"""
|
100 |
+
|
101 |
+
train_data = data["train"]
|
102 |
+
val_data = data["validation"]
|
103 |
+
|
104 |
+
#train_data = train_data.map(lambda x: tokenize(generate_prompt(x), tokenizer))
|
105 |
+
train_data = train_data_reduced.map(lambda x: tokenize(generate_prompt(x), tokenizer)) # use reduced train set
|
106 |
+
val_data = val_data.map(lambda x: tokenize(generate_prompt(x), tokenizer))
|
107 |
+
|
108 |
+
from transformers import AutoModelForCausalLM
|
109 |
+
|
110 |
+
print("Loading model for model: ", model_name[0])
|
111 |
+
|
112 |
+
model = AutoModelForCausalLM.from_pretrained(
|
113 |
+
afs_path+model_name[0],
|
114 |
+
load_in_8bit=False, # changed from True to False
|
115 |
+
device_map="auto",
|
116 |
+
)
|
117 |
+
|
118 |
+
|
119 |
+
|
120 |
+
# Define LoRA Config
|
121 |
+
lora_config = LoraConfig(
|
122 |
+
r= 8,
|
123 |
+
lora_alpha=16,
|
124 |
+
target_modules=["query_key_value"],
|
125 |
+
lora_dropout=0.05,
|
126 |
+
bias="none",
|
127 |
+
task_type=TaskType.CAUSAL_LM
|
128 |
+
)
|
129 |
+
|
130 |
+
|
131 |
+
# prepare int-8 model for training
|
132 |
+
#model = prepare_model_for_int8_training(model) #uncomment for int8
|
133 |
+
|
134 |
+
# add LoRA adaptor
|
135 |
+
model = get_peft_model(model, lora_config)
|
136 |
+
|
137 |
+
eval_steps = 200
|
138 |
+
save_steps = 200
|
139 |
+
logging_steps = 20
|
140 |
+
|
141 |
+
trainer = transformers.Trainer(
|
142 |
+
model=model,
|
143 |
+
train_dataset=train_data,
|
144 |
+
eval_dataset=val_data,
|
145 |
+
args=transformers.TrainingArguments(
|
146 |
+
num_train_epochs=3,
|
147 |
+
learning_rate=3e-4,
|
148 |
+
logging_steps=logging_steps,
|
149 |
+
logging_dir='./results', # directory for storing logs
|
150 |
+
evaluation_strategy="steps",
|
151 |
+
save_strategy="steps",
|
152 |
+
eval_steps=eval_steps,
|
153 |
+
save_steps=save_steps,
|
154 |
+
output_dir=output_dir,
|
155 |
+
report_to="none", #changed from report_to if report_to else to "none"
|
156 |
+
save_total_limit=3,
|
157 |
+
load_best_model_at_end=True,
|
158 |
+
push_to_hub=False,
|
159 |
+
auto_find_batch_size=True
|
160 |
+
),
|
161 |
+
data_collator=transformers.DataCollatorForLanguageModeling(tokenizer, mlm=False),
|
162 |
+
)
|
163 |
+
|
164 |
+
model.config.use_cache = False # silence the warnings. Please re-enable for inference!
|
165 |
+
|
166 |
+
trainer.train()
|
167 |
+
|
168 |
+
# Save our LoRA model & tokenizer results
|
169 |
+
trainer.model.save_pretrained(peft_name)
|
170 |
+
tokenizer.save_pretrained(peft_name)
|