End of training
Browse files- README.md +11 -0
- all_results.json +19 -0
- eval_results.json +13 -0
- generation_config.json +3 -3
- runs/Apr24_11-31-47_52ef395708b3/events.out.tfevents.1713991867.52ef395708b3.4125.1 +3 -0
- train_results.json +9 -0
- trainer_state.json +198 -0
README.md
CHANGED
@@ -1,7 +1,11 @@
|
|
1 |
---
|
|
|
|
|
2 |
base_model: Auruncus/lsg-lsh-mbart-cc25-16384-128-4-2ep
|
3 |
tags:
|
4 |
- generated_from_trainer
|
|
|
|
|
5 |
model-index:
|
6 |
- name: lsg-lsh-mbart-cc25-16384-128-4-4ep
|
7 |
results: []
|
@@ -13,6 +17,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
13 |
# lsg-lsh-mbart-cc25-16384-128-4-4ep
|
14 |
|
15 |
This model is a fine-tuned version of [Auruncus/lsg-lsh-mbart-cc25-16384-128-4-2ep](https://huggingface.co/Auruncus/lsg-lsh-mbart-cc25-16384-128-4-2ep) on an unknown dataset.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
16 |
|
17 |
## Model description
|
18 |
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- ru
|
4 |
base_model: Auruncus/lsg-lsh-mbart-cc25-16384-128-4-2ep
|
5 |
tags:
|
6 |
- generated_from_trainer
|
7 |
+
metrics:
|
8 |
+
- rouge
|
9 |
model-index:
|
10 |
- name: lsg-lsh-mbart-cc25-16384-128-4-4ep
|
11 |
results: []
|
|
|
17 |
# lsg-lsh-mbart-cc25-16384-128-4-4ep
|
18 |
|
19 |
This model is a fine-tuned version of [Auruncus/lsg-lsh-mbart-cc25-16384-128-4-2ep](https://huggingface.co/Auruncus/lsg-lsh-mbart-cc25-16384-128-4-2ep) on an unknown dataset.
|
20 |
+
It achieves the following results on the evaluation set:
|
21 |
+
- Loss: 2.5736
|
22 |
+
- Rouge1: 2.6972
|
23 |
+
- Rouge2: 1.6893
|
24 |
+
- Rougel: 2.6528
|
25 |
+
- Rougelsum: 2.6347
|
26 |
+
- Gen Len: 135.5
|
27 |
|
28 |
## Model description
|
29 |
|
all_results.json
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.9998339008388006,
|
3 |
+
"eval_gen_len": 135.5,
|
4 |
+
"eval_loss": 2.5736236572265625,
|
5 |
+
"eval_rouge1": 2.6972,
|
6 |
+
"eval_rouge2": 1.6893,
|
7 |
+
"eval_rougeL": 2.6528,
|
8 |
+
"eval_rougeLsum": 2.6347,
|
9 |
+
"eval_runtime": 504.2724,
|
10 |
+
"eval_samples": 200,
|
11 |
+
"eval_samples_per_second": 0.397,
|
12 |
+
"eval_steps_per_second": 0.397,
|
13 |
+
"total_flos": 1.2491757048924733e+18,
|
14 |
+
"train_loss": 0.2651168168185161,
|
15 |
+
"train_runtime": 32469.1467,
|
16 |
+
"train_samples": 48163,
|
17 |
+
"train_samples_per_second": 2.967,
|
18 |
+
"train_steps_per_second": 0.371
|
19 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.9998339008388006,
|
3 |
+
"eval_gen_len": 135.5,
|
4 |
+
"eval_loss": 2.5736236572265625,
|
5 |
+
"eval_rouge1": 2.6972,
|
6 |
+
"eval_rouge2": 1.6893,
|
7 |
+
"eval_rougeL": 2.6528,
|
8 |
+
"eval_rougeLsum": 2.6347,
|
9 |
+
"eval_runtime": 504.2724,
|
10 |
+
"eval_samples": 200,
|
11 |
+
"eval_samples_per_second": 0.397,
|
12 |
+
"eval_steps_per_second": 0.397
|
13 |
+
}
|
generation_config.json
CHANGED
@@ -1,10 +1,10 @@
|
|
1 |
{
|
2 |
-
"_from_model_config": true,
|
3 |
"bos_token_id": 0,
|
|
|
4 |
"eos_token_id": 2,
|
5 |
"forced_eos_token_id": 2,
|
6 |
-
"max_length":
|
7 |
-
"num_beams":
|
8 |
"pad_token_id": 1,
|
9 |
"transformers_version": "4.41.0.dev0"
|
10 |
}
|
|
|
1 |
{
|
|
|
2 |
"bos_token_id": 0,
|
3 |
+
"decoder_start_token_id": 250021,
|
4 |
"eos_token_id": 2,
|
5 |
"forced_eos_token_id": 2,
|
6 |
+
"max_length": 1024,
|
7 |
+
"num_beams": 5,
|
8 |
"pad_token_id": 1,
|
9 |
"transformers_version": "4.41.0.dev0"
|
10 |
}
|
runs/Apr24_11-31-47_52ef395708b3/events.out.tfevents.1713991867.52ef395708b3.4125.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:62ba85eaa01ce59f7ef12eec88ed2d8af13406a589836743c77f607d4de65eb4
|
3 |
+
size 613
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.9998339008388006,
|
3 |
+
"total_flos": 1.2491757048924733e+18,
|
4 |
+
"train_loss": 0.2651168168185161,
|
5 |
+
"train_runtime": 32469.1467,
|
6 |
+
"train_samples": 48163,
|
7 |
+
"train_samples_per_second": 2.967,
|
8 |
+
"train_steps_per_second": 0.371
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,198 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.9998339008388006,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 12040,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.08304958059961798,
|
13 |
+
"grad_norm": 0.3119010627269745,
|
14 |
+
"learning_rate": 4.7940199335548175e-05,
|
15 |
+
"loss": 0.3439,
|
16 |
+
"step": 500
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.16609916119923596,
|
20 |
+
"grad_norm": 0.9808832406997681,
|
21 |
+
"learning_rate": 4.5863787375415286e-05,
|
22 |
+
"loss": 0.2923,
|
23 |
+
"step": 1000
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.2491487417988539,
|
27 |
+
"grad_norm": 0.4935096800327301,
|
28 |
+
"learning_rate": 4.379568106312292e-05,
|
29 |
+
"loss": 0.4114,
|
30 |
+
"step": 1500
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.3321983223984719,
|
34 |
+
"grad_norm": 0.8764172196388245,
|
35 |
+
"learning_rate": 4.171926910299003e-05,
|
36 |
+
"loss": 0.309,
|
37 |
+
"step": 2000
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.41524790299808984,
|
41 |
+
"grad_norm": 0.9836592674255371,
|
42 |
+
"learning_rate": 3.964285714285714e-05,
|
43 |
+
"loss": 0.323,
|
44 |
+
"step": 2500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.4982974835977078,
|
48 |
+
"grad_norm": 0.8834519386291504,
|
49 |
+
"learning_rate": 3.7566445182724254e-05,
|
50 |
+
"loss": 0.2963,
|
51 |
+
"step": 3000
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.5813470641973258,
|
55 |
+
"grad_norm": 0.0723366066813469,
|
56 |
+
"learning_rate": 3.5490033222591366e-05,
|
57 |
+
"loss": 0.2738,
|
58 |
+
"step": 3500
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.6643966447969438,
|
62 |
+
"grad_norm": 0.799544095993042,
|
63 |
+
"learning_rate": 3.341362126245847e-05,
|
64 |
+
"loss": 0.2729,
|
65 |
+
"step": 4000
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.7474462253965617,
|
69 |
+
"grad_norm": 0.26322752237319946,
|
70 |
+
"learning_rate": 3.133720930232558e-05,
|
71 |
+
"loss": 0.2726,
|
72 |
+
"step": 4500
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.8304958059961797,
|
76 |
+
"grad_norm": 1.7207447290420532,
|
77 |
+
"learning_rate": 2.9264950166112958e-05,
|
78 |
+
"loss": 0.2854,
|
79 |
+
"step": 5000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.9135453865957976,
|
83 |
+
"grad_norm": 0.27596575021743774,
|
84 |
+
"learning_rate": 2.718853820598007e-05,
|
85 |
+
"loss": 0.2919,
|
86 |
+
"step": 5500
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.9965949671954156,
|
90 |
+
"grad_norm": 0.3219141364097595,
|
91 |
+
"learning_rate": 2.5112126245847177e-05,
|
92 |
+
"loss": 0.2948,
|
93 |
+
"step": 6000
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 1.0796445477950336,
|
97 |
+
"grad_norm": 1.064741849899292,
|
98 |
+
"learning_rate": 2.3035714285714285e-05,
|
99 |
+
"loss": 0.2182,
|
100 |
+
"step": 6500
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 1.1626941283946517,
|
104 |
+
"grad_norm": 1.5456783771514893,
|
105 |
+
"learning_rate": 2.0959302325581396e-05,
|
106 |
+
"loss": 0.1974,
|
107 |
+
"step": 7000
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 1.2457437089942696,
|
111 |
+
"grad_norm": 1.0485806465148926,
|
112 |
+
"learning_rate": 1.8882890365448504e-05,
|
113 |
+
"loss": 0.2068,
|
114 |
+
"step": 7500
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 1.3287932895938876,
|
118 |
+
"grad_norm": 0.9721829891204834,
|
119 |
+
"learning_rate": 1.6806478405315615e-05,
|
120 |
+
"loss": 0.211,
|
121 |
+
"step": 8000
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 1.4118428701935055,
|
125 |
+
"grad_norm": 0.15295402705669403,
|
126 |
+
"learning_rate": 1.4730066445182724e-05,
|
127 |
+
"loss": 0.2258,
|
128 |
+
"step": 8500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 1.4948924507931234,
|
132 |
+
"grad_norm": 1.4515725374221802,
|
133 |
+
"learning_rate": 1.2657807308970101e-05,
|
134 |
+
"loss": 0.2239,
|
135 |
+
"step": 9000
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 1.5779420313927415,
|
139 |
+
"grad_norm": 1.0659480094909668,
|
140 |
+
"learning_rate": 1.058139534883721e-05,
|
141 |
+
"loss": 0.2222,
|
142 |
+
"step": 9500
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 1.6609916119923596,
|
146 |
+
"grad_norm": 2.279104232788086,
|
147 |
+
"learning_rate": 8.50498338870432e-06,
|
148 |
+
"loss": 0.232,
|
149 |
+
"step": 10000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 1.7440411925919774,
|
153 |
+
"grad_norm": 0.7458949089050293,
|
154 |
+
"learning_rate": 6.428571428571429e-06,
|
155 |
+
"loss": 0.2378,
|
156 |
+
"step": 10500
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 1.8270907731915953,
|
160 |
+
"grad_norm": 1.7988941669464111,
|
161 |
+
"learning_rate": 4.352159468438539e-06,
|
162 |
+
"loss": 0.223,
|
163 |
+
"step": 11000
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 1.9101403537912134,
|
167 |
+
"grad_norm": 0.9544153213500977,
|
168 |
+
"learning_rate": 2.275747508305648e-06,
|
169 |
+
"loss": 0.2448,
|
170 |
+
"step": 11500
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 1.9931899343908315,
|
174 |
+
"grad_norm": 2.081665277481079,
|
175 |
+
"learning_rate": 1.9933554817275746e-07,
|
176 |
+
"loss": 0.2509,
|
177 |
+
"step": 12000
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 1.9998339008388006,
|
181 |
+
"step": 12040,
|
182 |
+
"total_flos": 1.2491757048924733e+18,
|
183 |
+
"train_loss": 0.2651168168185161,
|
184 |
+
"train_runtime": 32469.1467,
|
185 |
+
"train_samples_per_second": 2.967,
|
186 |
+
"train_steps_per_second": 0.371
|
187 |
+
}
|
188 |
+
],
|
189 |
+
"logging_steps": 500,
|
190 |
+
"max_steps": 12040,
|
191 |
+
"num_input_tokens_seen": 0,
|
192 |
+
"num_train_epochs": 2,
|
193 |
+
"save_steps": 500,
|
194 |
+
"total_flos": 1.2491757048924733e+18,
|
195 |
+
"train_batch_size": 2,
|
196 |
+
"trial_name": null,
|
197 |
+
"trial_params": null
|
198 |
+
}
|