Upload 10 files
Browse files- all_results.json +18 -0
- config.json +32 -0
- eval_results.json +11 -0
- generation_config.json +6 -0
- model.safetensors +3 -0
- special_tokens_map.json +23 -0
- tokenizer.json +0 -0
- tokenizer_config.json +214 -0
- train_results.json +11 -0
- trainer_state.json +1104 -0
all_results.json
ADDED
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 50.0,
|
3 |
+
"eval_accuracy": 0.2727662435089292,
|
4 |
+
"eval_loss": 4.3424248695373535,
|
5 |
+
"eval_runtime": 2.4316,
|
6 |
+
"eval_samples": 301,
|
7 |
+
"eval_samples_per_second": 123.785,
|
8 |
+
"eval_steps_per_second": 15.627,
|
9 |
+
"num_input_tokens_seen": 290048000,
|
10 |
+
"perplexity": 76.89377081598765,
|
11 |
+
"total_flos": 7.7740363481088e+16,
|
12 |
+
"train_loss": 3.9178111260633375,
|
13 |
+
"train_runtime": 4788.9263,
|
14 |
+
"train_samples": 5665,
|
15 |
+
"train_samples_per_second": 59.147,
|
16 |
+
"train_steps_per_second": 7.402,
|
17 |
+
"train_tokens_per_second": 60566.395
|
18 |
+
}
|
config.json
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "EleutherAI/pythia-70m-deduped",
|
3 |
+
"architectures": [
|
4 |
+
"GPTNeoXForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": true,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"bos_token_id": 0,
|
9 |
+
"classifier_dropout": 0.1,
|
10 |
+
"eos_token_id": 0,
|
11 |
+
"hidden_act": "gelu",
|
12 |
+
"hidden_dropout": 0.0,
|
13 |
+
"hidden_size": 512,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 2048,
|
16 |
+
"layer_norm_eps": 1e-05,
|
17 |
+
"max_position_embeddings": 2048,
|
18 |
+
"model_type": "gpt_neox",
|
19 |
+
"num_attention_heads": 8,
|
20 |
+
"num_hidden_layers": 6,
|
21 |
+
"partial_rotary_factor": 0.25,
|
22 |
+
"rope_scaling": null,
|
23 |
+
"rope_theta": 10000,
|
24 |
+
"rotary_emb_base": 10000,
|
25 |
+
"rotary_pct": 0.25,
|
26 |
+
"tie_word_embeddings": false,
|
27 |
+
"torch_dtype": "float32",
|
28 |
+
"transformers_version": "4.45.2",
|
29 |
+
"use_cache": true,
|
30 |
+
"use_parallel_residual": true,
|
31 |
+
"vocab_size": 50304
|
32 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 50.0,
|
3 |
+
"eval_accuracy": 0.2727662435089292,
|
4 |
+
"eval_loss": 4.3424248695373535,
|
5 |
+
"eval_runtime": 2.4316,
|
6 |
+
"eval_samples": 301,
|
7 |
+
"eval_samples_per_second": 123.785,
|
8 |
+
"eval_steps_per_second": 15.627,
|
9 |
+
"num_input_tokens_seen": 290048000,
|
10 |
+
"perplexity": 76.89377081598765
|
11 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 0,
|
4 |
+
"eos_token_id": 0,
|
5 |
+
"transformers_version": "4.45.2"
|
6 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8806c89e15bd19464e9f3700e946598c75501bc25a4f1b21f8b88fc86e46a04c
|
3 |
+
size 281715176
|
special_tokens_map.json
ADDED
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<|endoftext|>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"eos_token": {
|
10 |
+
"content": "<|endoftext|>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"unk_token": {
|
17 |
+
"content": "<|endoftext|>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
}
|
23 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,214 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_bos_token": false,
|
3 |
+
"add_eos_token": false,
|
4 |
+
"add_prefix_space": false,
|
5 |
+
"added_tokens_decoder": {
|
6 |
+
"0": {
|
7 |
+
"content": "<|endoftext|>",
|
8 |
+
"lstrip": false,
|
9 |
+
"normalized": false,
|
10 |
+
"rstrip": false,
|
11 |
+
"single_word": false,
|
12 |
+
"special": true
|
13 |
+
},
|
14 |
+
"1": {
|
15 |
+
"content": "<|padding|>",
|
16 |
+
"lstrip": false,
|
17 |
+
"normalized": false,
|
18 |
+
"rstrip": false,
|
19 |
+
"single_word": false,
|
20 |
+
"special": true
|
21 |
+
},
|
22 |
+
"50254": {
|
23 |
+
"content": " ",
|
24 |
+
"lstrip": false,
|
25 |
+
"normalized": true,
|
26 |
+
"rstrip": false,
|
27 |
+
"single_word": false,
|
28 |
+
"special": false
|
29 |
+
},
|
30 |
+
"50255": {
|
31 |
+
"content": " ",
|
32 |
+
"lstrip": false,
|
33 |
+
"normalized": true,
|
34 |
+
"rstrip": false,
|
35 |
+
"single_word": false,
|
36 |
+
"special": false
|
37 |
+
},
|
38 |
+
"50256": {
|
39 |
+
"content": " ",
|
40 |
+
"lstrip": false,
|
41 |
+
"normalized": true,
|
42 |
+
"rstrip": false,
|
43 |
+
"single_word": false,
|
44 |
+
"special": false
|
45 |
+
},
|
46 |
+
"50257": {
|
47 |
+
"content": " ",
|
48 |
+
"lstrip": false,
|
49 |
+
"normalized": true,
|
50 |
+
"rstrip": false,
|
51 |
+
"single_word": false,
|
52 |
+
"special": false
|
53 |
+
},
|
54 |
+
"50258": {
|
55 |
+
"content": " ",
|
56 |
+
"lstrip": false,
|
57 |
+
"normalized": true,
|
58 |
+
"rstrip": false,
|
59 |
+
"single_word": false,
|
60 |
+
"special": false
|
61 |
+
},
|
62 |
+
"50259": {
|
63 |
+
"content": " ",
|
64 |
+
"lstrip": false,
|
65 |
+
"normalized": true,
|
66 |
+
"rstrip": false,
|
67 |
+
"single_word": false,
|
68 |
+
"special": false
|
69 |
+
},
|
70 |
+
"50260": {
|
71 |
+
"content": " ",
|
72 |
+
"lstrip": false,
|
73 |
+
"normalized": true,
|
74 |
+
"rstrip": false,
|
75 |
+
"single_word": false,
|
76 |
+
"special": false
|
77 |
+
},
|
78 |
+
"50261": {
|
79 |
+
"content": " ",
|
80 |
+
"lstrip": false,
|
81 |
+
"normalized": true,
|
82 |
+
"rstrip": false,
|
83 |
+
"single_word": false,
|
84 |
+
"special": false
|
85 |
+
},
|
86 |
+
"50262": {
|
87 |
+
"content": " ",
|
88 |
+
"lstrip": false,
|
89 |
+
"normalized": true,
|
90 |
+
"rstrip": false,
|
91 |
+
"single_word": false,
|
92 |
+
"special": false
|
93 |
+
},
|
94 |
+
"50263": {
|
95 |
+
"content": " ",
|
96 |
+
"lstrip": false,
|
97 |
+
"normalized": true,
|
98 |
+
"rstrip": false,
|
99 |
+
"single_word": false,
|
100 |
+
"special": false
|
101 |
+
},
|
102 |
+
"50264": {
|
103 |
+
"content": " ",
|
104 |
+
"lstrip": false,
|
105 |
+
"normalized": true,
|
106 |
+
"rstrip": false,
|
107 |
+
"single_word": false,
|
108 |
+
"special": false
|
109 |
+
},
|
110 |
+
"50265": {
|
111 |
+
"content": " ",
|
112 |
+
"lstrip": false,
|
113 |
+
"normalized": true,
|
114 |
+
"rstrip": false,
|
115 |
+
"single_word": false,
|
116 |
+
"special": false
|
117 |
+
},
|
118 |
+
"50266": {
|
119 |
+
"content": " ",
|
120 |
+
"lstrip": false,
|
121 |
+
"normalized": true,
|
122 |
+
"rstrip": false,
|
123 |
+
"single_word": false,
|
124 |
+
"special": false
|
125 |
+
},
|
126 |
+
"50267": {
|
127 |
+
"content": " ",
|
128 |
+
"lstrip": false,
|
129 |
+
"normalized": true,
|
130 |
+
"rstrip": false,
|
131 |
+
"single_word": false,
|
132 |
+
"special": false
|
133 |
+
},
|
134 |
+
"50268": {
|
135 |
+
"content": " ",
|
136 |
+
"lstrip": false,
|
137 |
+
"normalized": true,
|
138 |
+
"rstrip": false,
|
139 |
+
"single_word": false,
|
140 |
+
"special": false
|
141 |
+
},
|
142 |
+
"50269": {
|
143 |
+
"content": " ",
|
144 |
+
"lstrip": false,
|
145 |
+
"normalized": true,
|
146 |
+
"rstrip": false,
|
147 |
+
"single_word": false,
|
148 |
+
"special": false
|
149 |
+
},
|
150 |
+
"50270": {
|
151 |
+
"content": " ",
|
152 |
+
"lstrip": false,
|
153 |
+
"normalized": true,
|
154 |
+
"rstrip": false,
|
155 |
+
"single_word": false,
|
156 |
+
"special": false
|
157 |
+
},
|
158 |
+
"50271": {
|
159 |
+
"content": " ",
|
160 |
+
"lstrip": false,
|
161 |
+
"normalized": true,
|
162 |
+
"rstrip": false,
|
163 |
+
"single_word": false,
|
164 |
+
"special": false
|
165 |
+
},
|
166 |
+
"50272": {
|
167 |
+
"content": " ",
|
168 |
+
"lstrip": false,
|
169 |
+
"normalized": true,
|
170 |
+
"rstrip": false,
|
171 |
+
"single_word": false,
|
172 |
+
"special": false
|
173 |
+
},
|
174 |
+
"50273": {
|
175 |
+
"content": " ",
|
176 |
+
"lstrip": false,
|
177 |
+
"normalized": true,
|
178 |
+
"rstrip": false,
|
179 |
+
"single_word": false,
|
180 |
+
"special": false
|
181 |
+
},
|
182 |
+
"50274": {
|
183 |
+
"content": " ",
|
184 |
+
"lstrip": false,
|
185 |
+
"normalized": true,
|
186 |
+
"rstrip": false,
|
187 |
+
"single_word": false,
|
188 |
+
"special": false
|
189 |
+
},
|
190 |
+
"50275": {
|
191 |
+
"content": " ",
|
192 |
+
"lstrip": false,
|
193 |
+
"normalized": true,
|
194 |
+
"rstrip": false,
|
195 |
+
"single_word": false,
|
196 |
+
"special": false
|
197 |
+
},
|
198 |
+
"50276": {
|
199 |
+
"content": " ",
|
200 |
+
"lstrip": false,
|
201 |
+
"normalized": true,
|
202 |
+
"rstrip": false,
|
203 |
+
"single_word": false,
|
204 |
+
"special": false
|
205 |
+
}
|
206 |
+
},
|
207 |
+
"bos_token": "<|endoftext|>",
|
208 |
+
"clean_up_tokenization_spaces": false,
|
209 |
+
"eos_token": "<|endoftext|>",
|
210 |
+
"model_max_length": 1000000000000000019884624838656,
|
211 |
+
"pad_token": null,
|
212 |
+
"tokenizer_class": "GPTNeoXTokenizer",
|
213 |
+
"unk_token": "<|endoftext|>"
|
214 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 50.0,
|
3 |
+
"num_input_tokens_seen": 290048000,
|
4 |
+
"total_flos": 7.7740363481088e+16,
|
5 |
+
"train_loss": 3.9178111260633375,
|
6 |
+
"train_runtime": 4788.9263,
|
7 |
+
"train_samples": 5665,
|
8 |
+
"train_samples_per_second": 59.147,
|
9 |
+
"train_steps_per_second": 7.402,
|
10 |
+
"train_tokens_per_second": 60566.395
|
11 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1104 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 4.3424248695373535,
|
3 |
+
"best_model_checkpoint": "/media/user/Expansion/pythia-70m-wikipedia-paragraphs-50epochs/checkpoint-11344",
|
4 |
+
"epoch": 50.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 35450,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.7052186177715092,
|
13 |
+
"grad_norm": 84.19266510009766,
|
14 |
+
"learning_rate": 4.929478138222849e-05,
|
15 |
+
"loss": 7.2253,
|
16 |
+
"num_input_tokens_seen": 4096000,
|
17 |
+
"step": 500
|
18 |
+
},
|
19 |
+
{
|
20 |
+
"epoch": 1.0,
|
21 |
+
"eval_accuracy": 0.1339588143789194,
|
22 |
+
"eval_loss": 6.6662068367004395,
|
23 |
+
"eval_runtime": 2.3283,
|
24 |
+
"eval_samples_per_second": 129.28,
|
25 |
+
"eval_steps_per_second": 16.321,
|
26 |
+
"num_input_tokens_seen": 5800960,
|
27 |
+
"step": 709
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 1.4104372355430184,
|
31 |
+
"grad_norm": 34.27415466308594,
|
32 |
+
"learning_rate": 4.858956276445699e-05,
|
33 |
+
"loss": 6.5693,
|
34 |
+
"num_input_tokens_seen": 8184832,
|
35 |
+
"step": 1000
|
36 |
+
},
|
37 |
+
{
|
38 |
+
"epoch": 2.0,
|
39 |
+
"eval_accuracy": 0.13123735479324378,
|
40 |
+
"eval_loss": 6.502222061157227,
|
41 |
+
"eval_runtime": 2.5457,
|
42 |
+
"eval_samples_per_second": 118.237,
|
43 |
+
"eval_steps_per_second": 14.927,
|
44 |
+
"num_input_tokens_seen": 11601920,
|
45 |
+
"step": 1418
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 2.1156558533145273,
|
49 |
+
"grad_norm": 62.75070571899414,
|
50 |
+
"learning_rate": 4.788434414668547e-05,
|
51 |
+
"loss": 6.235,
|
52 |
+
"num_input_tokens_seen": 12273664,
|
53 |
+
"step": 1500
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 2.8208744710860367,
|
57 |
+
"grad_norm": 83.26922607421875,
|
58 |
+
"learning_rate": 4.717912552891397e-05,
|
59 |
+
"loss": 5.9786,
|
60 |
+
"num_input_tokens_seen": 16369664,
|
61 |
+
"step": 2000
|
62 |
+
},
|
63 |
+
{
|
64 |
+
"epoch": 3.0,
|
65 |
+
"eval_accuracy": 0.15382417032829637,
|
66 |
+
"eval_loss": 6.0094099044799805,
|
67 |
+
"eval_runtime": 2.2872,
|
68 |
+
"eval_samples_per_second": 131.601,
|
69 |
+
"eval_steps_per_second": 16.614,
|
70 |
+
"num_input_tokens_seen": 17402880,
|
71 |
+
"step": 2127
|
72 |
+
},
|
73 |
+
{
|
74 |
+
"epoch": 3.5260930888575457,
|
75 |
+
"grad_norm": 22.89128875732422,
|
76 |
+
"learning_rate": 4.647390691114246e-05,
|
77 |
+
"loss": 5.3855,
|
78 |
+
"num_input_tokens_seen": 20458496,
|
79 |
+
"step": 2500
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 4.0,
|
83 |
+
"eval_accuracy": 0.19935828112872375,
|
84 |
+
"eval_loss": 5.0559468269348145,
|
85 |
+
"eval_runtime": 2.3199,
|
86 |
+
"eval_samples_per_second": 129.746,
|
87 |
+
"eval_steps_per_second": 16.38,
|
88 |
+
"num_input_tokens_seen": 23203840,
|
89 |
+
"step": 2836
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 4.231311706629055,
|
93 |
+
"grad_norm": 45.5040168762207,
|
94 |
+
"learning_rate": 4.576868829337095e-05,
|
95 |
+
"loss": 5.0498,
|
96 |
+
"num_input_tokens_seen": 24547328,
|
97 |
+
"step": 3000
|
98 |
+
},
|
99 |
+
{
|
100 |
+
"epoch": 4.9365303244005645,
|
101 |
+
"grad_norm": 33.329322814941406,
|
102 |
+
"learning_rate": 4.506346967559944e-05,
|
103 |
+
"loss": 4.9003,
|
104 |
+
"num_input_tokens_seen": 28643328,
|
105 |
+
"step": 3500
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 5.0,
|
109 |
+
"eval_accuracy": 0.20717517041598063,
|
110 |
+
"eval_loss": 4.983646869659424,
|
111 |
+
"eval_runtime": 2.3314,
|
112 |
+
"eval_samples_per_second": 129.105,
|
113 |
+
"eval_steps_per_second": 16.299,
|
114 |
+
"num_input_tokens_seen": 29004800,
|
115 |
+
"step": 3545
|
116 |
+
},
|
117 |
+
{
|
118 |
+
"epoch": 5.6417489421720735,
|
119 |
+
"grad_norm": 23.64671516418457,
|
120 |
+
"learning_rate": 4.4358251057827926e-05,
|
121 |
+
"loss": 4.726,
|
122 |
+
"num_input_tokens_seen": 32732160,
|
123 |
+
"step": 4000
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 6.0,
|
127 |
+
"eval_accuracy": 0.24000805396154234,
|
128 |
+
"eval_loss": 4.574463844299316,
|
129 |
+
"eval_runtime": 2.395,
|
130 |
+
"eval_samples_per_second": 125.68,
|
131 |
+
"eval_steps_per_second": 15.867,
|
132 |
+
"num_input_tokens_seen": 34805760,
|
133 |
+
"step": 4254
|
134 |
+
},
|
135 |
+
{
|
136 |
+
"epoch": 6.346967559943582,
|
137 |
+
"grad_norm": 64.83744049072266,
|
138 |
+
"learning_rate": 4.3653032440056416e-05,
|
139 |
+
"loss": 4.5269,
|
140 |
+
"num_input_tokens_seen": 36820992,
|
141 |
+
"step": 4500
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 7.0,
|
145 |
+
"eval_accuracy": 0.2230005553336386,
|
146 |
+
"eval_loss": 4.76109504699707,
|
147 |
+
"eval_runtime": 2.3456,
|
148 |
+
"eval_samples_per_second": 128.328,
|
149 |
+
"eval_steps_per_second": 16.201,
|
150 |
+
"num_input_tokens_seen": 40606720,
|
151 |
+
"step": 4963
|
152 |
+
},
|
153 |
+
{
|
154 |
+
"epoch": 7.052186177715091,
|
155 |
+
"grad_norm": 24.155548095703125,
|
156 |
+
"learning_rate": 4.294781382228491e-05,
|
157 |
+
"loss": 4.5745,
|
158 |
+
"num_input_tokens_seen": 40909824,
|
159 |
+
"step": 5000
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 7.757404795486601,
|
163 |
+
"grad_norm": 16.52519416809082,
|
164 |
+
"learning_rate": 4.22425952045134e-05,
|
165 |
+
"loss": 4.5249,
|
166 |
+
"num_input_tokens_seen": 45005824,
|
167 |
+
"step": 5500
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 8.0,
|
171 |
+
"eval_accuracy": 0.23566605937198584,
|
172 |
+
"eval_loss": 4.611110687255859,
|
173 |
+
"eval_runtime": 2.3827,
|
174 |
+
"eval_samples_per_second": 126.326,
|
175 |
+
"eval_steps_per_second": 15.948,
|
176 |
+
"num_input_tokens_seen": 46407680,
|
177 |
+
"step": 5672
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 8.46262341325811,
|
181 |
+
"grad_norm": 19.4478759765625,
|
182 |
+
"learning_rate": 4.153737658674189e-05,
|
183 |
+
"loss": 4.3857,
|
184 |
+
"num_input_tokens_seen": 49094656,
|
185 |
+
"step": 6000
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 9.0,
|
189 |
+
"eval_accuracy": 0.24934155616826284,
|
190 |
+
"eval_loss": 4.472122669219971,
|
191 |
+
"eval_runtime": 2.3575,
|
192 |
+
"eval_samples_per_second": 127.679,
|
193 |
+
"eval_steps_per_second": 16.119,
|
194 |
+
"num_input_tokens_seen": 52208640,
|
195 |
+
"step": 6381
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 9.16784203102962,
|
199 |
+
"grad_norm": 12.421043395996094,
|
200 |
+
"learning_rate": 4.083215796897038e-05,
|
201 |
+
"loss": 4.2982,
|
202 |
+
"num_input_tokens_seen": 53183488,
|
203 |
+
"step": 6500
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 9.873060648801129,
|
207 |
+
"grad_norm": 19.52003288269043,
|
208 |
+
"learning_rate": 4.012693935119887e-05,
|
209 |
+
"loss": 4.2523,
|
210 |
+
"num_input_tokens_seen": 57279488,
|
211 |
+
"step": 7000
|
212 |
+
},
|
213 |
+
{
|
214 |
+
"epoch": 10.0,
|
215 |
+
"eval_accuracy": 0.2492506243443978,
|
216 |
+
"eval_loss": 4.475858688354492,
|
217 |
+
"eval_runtime": 2.3615,
|
218 |
+
"eval_samples_per_second": 127.463,
|
219 |
+
"eval_steps_per_second": 16.092,
|
220 |
+
"num_input_tokens_seen": 58009600,
|
221 |
+
"step": 7090
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 10.578279266572638,
|
225 |
+
"grad_norm": 13.172361373901367,
|
226 |
+
"learning_rate": 3.942172073342737e-05,
|
227 |
+
"loss": 4.2256,
|
228 |
+
"num_input_tokens_seen": 61368320,
|
229 |
+
"step": 7500
|
230 |
+
},
|
231 |
+
{
|
232 |
+
"epoch": 11.0,
|
233 |
+
"eval_accuracy": 0.2573857750151824,
|
234 |
+
"eval_loss": 4.4382004737854,
|
235 |
+
"eval_runtime": 2.3391,
|
236 |
+
"eval_samples_per_second": 128.68,
|
237 |
+
"eval_steps_per_second": 16.245,
|
238 |
+
"num_input_tokens_seen": 63810560,
|
239 |
+
"step": 7799
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 11.283497884344147,
|
243 |
+
"grad_norm": 20.001747131347656,
|
244 |
+
"learning_rate": 3.871650211565586e-05,
|
245 |
+
"loss": 4.151,
|
246 |
+
"num_input_tokens_seen": 65457152,
|
247 |
+
"step": 8000
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 11.988716502115656,
|
251 |
+
"grad_norm": 12.548118591308594,
|
252 |
+
"learning_rate": 3.801128349788435e-05,
|
253 |
+
"loss": 4.0897,
|
254 |
+
"num_input_tokens_seen": 69553152,
|
255 |
+
"step": 8500
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 12.0,
|
259 |
+
"eval_accuracy": 0.26011697729627214,
|
260 |
+
"eval_loss": 4.386884689331055,
|
261 |
+
"eval_runtime": 2.3571,
|
262 |
+
"eval_samples_per_second": 127.697,
|
263 |
+
"eval_steps_per_second": 16.121,
|
264 |
+
"num_input_tokens_seen": 69611520,
|
265 |
+
"step": 8508
|
266 |
+
},
|
267 |
+
{
|
268 |
+
"epoch": 12.693935119887165,
|
269 |
+
"grad_norm": 13.00560188293457,
|
270 |
+
"learning_rate": 3.7306064880112836e-05,
|
271 |
+
"loss": 4.0155,
|
272 |
+
"num_input_tokens_seen": 73641984,
|
273 |
+
"step": 9000
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 13.0,
|
277 |
+
"eval_accuracy": 0.2617082842139106,
|
278 |
+
"eval_loss": 4.3910322189331055,
|
279 |
+
"eval_runtime": 2.3687,
|
280 |
+
"eval_samples_per_second": 127.077,
|
281 |
+
"eval_steps_per_second": 16.043,
|
282 |
+
"num_input_tokens_seen": 75412480,
|
283 |
+
"step": 9217
|
284 |
+
},
|
285 |
+
{
|
286 |
+
"epoch": 13.399153737658674,
|
287 |
+
"grad_norm": 17.632902145385742,
|
288 |
+
"learning_rate": 3.6600846262341326e-05,
|
289 |
+
"loss": 4.0051,
|
290 |
+
"num_input_tokens_seen": 77730816,
|
291 |
+
"step": 9500
|
292 |
+
},
|
293 |
+
{
|
294 |
+
"epoch": 14.0,
|
295 |
+
"eval_accuracy": 0.2632378873939264,
|
296 |
+
"eval_loss": 4.3866143226623535,
|
297 |
+
"eval_runtime": 2.3505,
|
298 |
+
"eval_samples_per_second": 128.057,
|
299 |
+
"eval_steps_per_second": 16.167,
|
300 |
+
"num_input_tokens_seen": 81213440,
|
301 |
+
"step": 9926
|
302 |
+
},
|
303 |
+
{
|
304 |
+
"epoch": 14.104372355430183,
|
305 |
+
"grad_norm": 8.833736419677734,
|
306 |
+
"learning_rate": 3.5895627644569816e-05,
|
307 |
+
"loss": 3.9612,
|
308 |
+
"num_input_tokens_seen": 81819648,
|
309 |
+
"step": 10000
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 14.809590973201692,
|
313 |
+
"grad_norm": 12.257335662841797,
|
314 |
+
"learning_rate": 3.519040902679831e-05,
|
315 |
+
"loss": 3.9221,
|
316 |
+
"num_input_tokens_seen": 85915648,
|
317 |
+
"step": 10500
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 15.0,
|
321 |
+
"eval_accuracy": 0.2629553492269171,
|
322 |
+
"eval_loss": 4.392822742462158,
|
323 |
+
"eval_runtime": 2.3838,
|
324 |
+
"eval_samples_per_second": 126.269,
|
325 |
+
"eval_steps_per_second": 15.941,
|
326 |
+
"num_input_tokens_seen": 87014400,
|
327 |
+
"step": 10635
|
328 |
+
},
|
329 |
+
{
|
330 |
+
"epoch": 15.514809590973202,
|
331 |
+
"grad_norm": 10.67398738861084,
|
332 |
+
"learning_rate": 3.44851904090268e-05,
|
333 |
+
"loss": 3.8844,
|
334 |
+
"num_input_tokens_seen": 90004480,
|
335 |
+
"step": 11000
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 16.0,
|
339 |
+
"eval_accuracy": 0.2727662435089292,
|
340 |
+
"eval_loss": 4.3424248695373535,
|
341 |
+
"eval_runtime": 2.3672,
|
342 |
+
"eval_samples_per_second": 127.154,
|
343 |
+
"eval_steps_per_second": 16.053,
|
344 |
+
"num_input_tokens_seen": 92815360,
|
345 |
+
"step": 11344
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 16.22002820874471,
|
349 |
+
"grad_norm": 9.101444244384766,
|
350 |
+
"learning_rate": 3.377997179125529e-05,
|
351 |
+
"loss": 3.8528,
|
352 |
+
"num_input_tokens_seen": 94093312,
|
353 |
+
"step": 11500
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 16.92524682651622,
|
357 |
+
"grad_norm": 9.585593223571777,
|
358 |
+
"learning_rate": 3.307475317348378e-05,
|
359 |
+
"loss": 3.8165,
|
360 |
+
"num_input_tokens_seen": 98189312,
|
361 |
+
"step": 12000
|
362 |
+
},
|
363 |
+
{
|
364 |
+
"epoch": 17.0,
|
365 |
+
"eval_accuracy": 0.2663425596658905,
|
366 |
+
"eval_loss": 4.406184196472168,
|
367 |
+
"eval_runtime": 2.3686,
|
368 |
+
"eval_samples_per_second": 127.081,
|
369 |
+
"eval_steps_per_second": 16.043,
|
370 |
+
"num_input_tokens_seen": 98616320,
|
371 |
+
"step": 12053
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 17.63046544428773,
|
375 |
+
"grad_norm": 12.549637794494629,
|
376 |
+
"learning_rate": 3.236953455571227e-05,
|
377 |
+
"loss": 3.7716,
|
378 |
+
"num_input_tokens_seen": 102278144,
|
379 |
+
"step": 12500
|
380 |
+
},
|
381 |
+
{
|
382 |
+
"epoch": 18.0,
|
383 |
+
"eval_accuracy": 0.26832682196523155,
|
384 |
+
"eval_loss": 4.423529624938965,
|
385 |
+
"eval_runtime": 2.3586,
|
386 |
+
"eval_samples_per_second": 127.62,
|
387 |
+
"eval_steps_per_second": 16.111,
|
388 |
+
"num_input_tokens_seen": 104417280,
|
389 |
+
"step": 12762
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 18.33568406205924,
|
393 |
+
"grad_norm": 10.797262191772461,
|
394 |
+
"learning_rate": 3.166431593794077e-05,
|
395 |
+
"loss": 3.7558,
|
396 |
+
"num_input_tokens_seen": 106366976,
|
397 |
+
"step": 13000
|
398 |
+
},
|
399 |
+
{
|
400 |
+
"epoch": 19.0,
|
401 |
+
"eval_accuracy": 0.26874575786803845,
|
402 |
+
"eval_loss": 4.390767574310303,
|
403 |
+
"eval_runtime": 2.3418,
|
404 |
+
"eval_samples_per_second": 128.535,
|
405 |
+
"eval_steps_per_second": 16.227,
|
406 |
+
"num_input_tokens_seen": 110218240,
|
407 |
+
"step": 13471
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 19.040902679830747,
|
411 |
+
"grad_norm": 10.686503410339355,
|
412 |
+
"learning_rate": 3.095909732016926e-05,
|
413 |
+
"loss": 3.7383,
|
414 |
+
"num_input_tokens_seen": 110455808,
|
415 |
+
"step": 13500
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 19.746121297602258,
|
419 |
+
"grad_norm": 8.79494857788086,
|
420 |
+
"learning_rate": 3.0253878702397743e-05,
|
421 |
+
"loss": 3.6958,
|
422 |
+
"num_input_tokens_seen": 114551808,
|
423 |
+
"step": 14000
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 20.0,
|
427 |
+
"eval_accuracy": 0.26914196081487907,
|
428 |
+
"eval_loss": 4.424190521240234,
|
429 |
+
"eval_runtime": 2.4033,
|
430 |
+
"eval_samples_per_second": 125.243,
|
431 |
+
"eval_steps_per_second": 15.811,
|
432 |
+
"num_input_tokens_seen": 116019200,
|
433 |
+
"step": 14180
|
434 |
+
},
|
435 |
+
{
|
436 |
+
"epoch": 20.451339915373765,
|
437 |
+
"grad_norm": 8.323198318481445,
|
438 |
+
"learning_rate": 2.9548660084626233e-05,
|
439 |
+
"loss": 3.684,
|
440 |
+
"num_input_tokens_seen": 118640640,
|
441 |
+
"step": 14500
|
442 |
+
},
|
443 |
+
{
|
444 |
+
"epoch": 21.0,
|
445 |
+
"eval_accuracy": 0.26918742672681156,
|
446 |
+
"eval_loss": 4.429287433624268,
|
447 |
+
"eval_runtime": 2.3485,
|
448 |
+
"eval_samples_per_second": 128.164,
|
449 |
+
"eval_steps_per_second": 16.18,
|
450 |
+
"num_input_tokens_seen": 121820160,
|
451 |
+
"step": 14889
|
452 |
+
},
|
453 |
+
{
|
454 |
+
"epoch": 21.156558533145276,
|
455 |
+
"grad_norm": 13.749457359313965,
|
456 |
+
"learning_rate": 2.8843441466854726e-05,
|
457 |
+
"loss": 3.6566,
|
458 |
+
"num_input_tokens_seen": 122729472,
|
459 |
+
"step": 15000
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 21.861777150916783,
|
463 |
+
"grad_norm": 8.61412239074707,
|
464 |
+
"learning_rate": 2.813822284908322e-05,
|
465 |
+
"loss": 3.6349,
|
466 |
+
"num_input_tokens_seen": 126825472,
|
467 |
+
"step": 15500
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 22.0,
|
471 |
+
"eval_accuracy": 0.26987591053607557,
|
472 |
+
"eval_loss": 4.465507984161377,
|
473 |
+
"eval_runtime": 2.4176,
|
474 |
+
"eval_samples_per_second": 124.505,
|
475 |
+
"eval_steps_per_second": 15.718,
|
476 |
+
"num_input_tokens_seen": 127621120,
|
477 |
+
"step": 15598
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 22.566995768688294,
|
481 |
+
"grad_norm": 8.218836784362793,
|
482 |
+
"learning_rate": 2.743300423131171e-05,
|
483 |
+
"loss": 3.6091,
|
484 |
+
"num_input_tokens_seen": 130914304,
|
485 |
+
"step": 16000
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 23.0,
|
489 |
+
"eval_accuracy": 0.2670992423430533,
|
490 |
+
"eval_loss": 4.5008721351623535,
|
491 |
+
"eval_runtime": 2.366,
|
492 |
+
"eval_samples_per_second": 127.22,
|
493 |
+
"eval_steps_per_second": 16.061,
|
494 |
+
"num_input_tokens_seen": 133422080,
|
495 |
+
"step": 16307
|
496 |
+
},
|
497 |
+
{
|
498 |
+
"epoch": 23.2722143864598,
|
499 |
+
"grad_norm": 10.475564956665039,
|
500 |
+
"learning_rate": 2.6727785613540195e-05,
|
501 |
+
"loss": 3.602,
|
502 |
+
"num_input_tokens_seen": 135003136,
|
503 |
+
"step": 16500
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 23.97743300423131,
|
507 |
+
"grad_norm": 9.387394905090332,
|
508 |
+
"learning_rate": 2.6022566995768688e-05,
|
509 |
+
"loss": 3.5949,
|
510 |
+
"num_input_tokens_seen": 139099136,
|
511 |
+
"step": 17000
|
512 |
+
},
|
513 |
+
{
|
514 |
+
"epoch": 24.0,
|
515 |
+
"eval_accuracy": 0.26471877709687164,
|
516 |
+
"eval_loss": 4.58220911026001,
|
517 |
+
"eval_runtime": 2.3413,
|
518 |
+
"eval_samples_per_second": 128.561,
|
519 |
+
"eval_steps_per_second": 16.23,
|
520 |
+
"num_input_tokens_seen": 139223040,
|
521 |
+
"step": 17016
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 24.682651622002822,
|
525 |
+
"grad_norm": 9.367217063903809,
|
526 |
+
"learning_rate": 2.531734837799718e-05,
|
527 |
+
"loss": 3.5476,
|
528 |
+
"num_input_tokens_seen": 143187968,
|
529 |
+
"step": 17500
|
530 |
+
},
|
531 |
+
{
|
532 |
+
"epoch": 25.0,
|
533 |
+
"eval_accuracy": 0.2674499793779614,
|
534 |
+
"eval_loss": 4.5724711418151855,
|
535 |
+
"eval_runtime": 2.3479,
|
536 |
+
"eval_samples_per_second": 128.2,
|
537 |
+
"eval_steps_per_second": 16.185,
|
538 |
+
"num_input_tokens_seen": 145024000,
|
539 |
+
"step": 17725
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 25.38787023977433,
|
543 |
+
"grad_norm": 9.069378852844238,
|
544 |
+
"learning_rate": 2.461212976022567e-05,
|
545 |
+
"loss": 3.5392,
|
546 |
+
"num_input_tokens_seen": 147276800,
|
547 |
+
"step": 18000
|
548 |
+
},
|
549 |
+
{
|
550 |
+
"epoch": 26.0,
|
551 |
+
"eval_accuracy": 0.266644583223728,
|
552 |
+
"eval_loss": 4.611934185028076,
|
553 |
+
"eval_runtime": 2.372,
|
554 |
+
"eval_samples_per_second": 126.897,
|
555 |
+
"eval_steps_per_second": 16.02,
|
556 |
+
"num_input_tokens_seen": 150824960,
|
557 |
+
"step": 18434
|
558 |
+
},
|
559 |
+
{
|
560 |
+
"epoch": 26.09308885754584,
|
561 |
+
"grad_norm": 7.782838821411133,
|
562 |
+
"learning_rate": 2.390691114245416e-05,
|
563 |
+
"loss": 3.5423,
|
564 |
+
"num_input_tokens_seen": 151365632,
|
565 |
+
"step": 18500
|
566 |
+
},
|
567 |
+
{
|
568 |
+
"epoch": 26.798307475317348,
|
569 |
+
"grad_norm": 7.378510475158691,
|
570 |
+
"learning_rate": 2.3201692524682653e-05,
|
571 |
+
"loss": 3.524,
|
572 |
+
"num_input_tokens_seen": 155461632,
|
573 |
+
"step": 19000
|
574 |
+
},
|
575 |
+
{
|
576 |
+
"epoch": 27.0,
|
577 |
+
"eval_accuracy": 0.26156214378269893,
|
578 |
+
"eval_loss": 4.708793640136719,
|
579 |
+
"eval_runtime": 2.4833,
|
580 |
+
"eval_samples_per_second": 121.209,
|
581 |
+
"eval_steps_per_second": 15.302,
|
582 |
+
"num_input_tokens_seen": 156625920,
|
583 |
+
"step": 19143
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 27.50352609308886,
|
587 |
+
"grad_norm": 8.874760627746582,
|
588 |
+
"learning_rate": 2.2496473906911146e-05,
|
589 |
+
"loss": 3.4999,
|
590 |
+
"num_input_tokens_seen": 159550464,
|
591 |
+
"step": 19500
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 28.0,
|
595 |
+
"eval_accuracy": 0.26340676078110437,
|
596 |
+
"eval_loss": 4.730751037597656,
|
597 |
+
"eval_runtime": 2.3629,
|
598 |
+
"eval_samples_per_second": 127.388,
|
599 |
+
"eval_steps_per_second": 16.082,
|
600 |
+
"num_input_tokens_seen": 162426880,
|
601 |
+
"step": 19852
|
602 |
+
},
|
603 |
+
{
|
604 |
+
"epoch": 28.208744710860366,
|
605 |
+
"grad_norm": 8.382628440856934,
|
606 |
+
"learning_rate": 2.1791255289139633e-05,
|
607 |
+
"loss": 3.5073,
|
608 |
+
"num_input_tokens_seen": 163639296,
|
609 |
+
"step": 20000
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 28.913963328631876,
|
613 |
+
"grad_norm": 7.747939109802246,
|
614 |
+
"learning_rate": 2.1086036671368126e-05,
|
615 |
+
"loss": 3.4888,
|
616 |
+
"num_input_tokens_seen": 167735296,
|
617 |
+
"step": 20500
|
618 |
+
},
|
619 |
+
{
|
620 |
+
"epoch": 29.0,
|
621 |
+
"eval_accuracy": 0.26255589871493845,
|
622 |
+
"eval_loss": 4.776463031768799,
|
623 |
+
"eval_runtime": 2.3641,
|
624 |
+
"eval_samples_per_second": 127.324,
|
625 |
+
"eval_steps_per_second": 16.074,
|
626 |
+
"num_input_tokens_seen": 168227840,
|
627 |
+
"step": 20561
|
628 |
+
},
|
629 |
+
{
|
630 |
+
"epoch": 29.619181946403383,
|
631 |
+
"grad_norm": 10.21652603149414,
|
632 |
+
"learning_rate": 2.0380818053596615e-05,
|
633 |
+
"loss": 3.4665,
|
634 |
+
"num_input_tokens_seen": 171824128,
|
635 |
+
"step": 21000
|
636 |
+
},
|
637 |
+
{
|
638 |
+
"epoch": 30.0,
|
639 |
+
"eval_accuracy": 0.2627962185351533,
|
640 |
+
"eval_loss": 4.861045837402344,
|
641 |
+
"eval_runtime": 2.3278,
|
642 |
+
"eval_samples_per_second": 129.308,
|
643 |
+
"eval_steps_per_second": 16.325,
|
644 |
+
"num_input_tokens_seen": 174028800,
|
645 |
+
"step": 21270
|
646 |
+
},
|
647 |
+
{
|
648 |
+
"epoch": 30.324400564174894,
|
649 |
+
"grad_norm": 7.920512676239014,
|
650 |
+
"learning_rate": 1.9675599435825108e-05,
|
651 |
+
"loss": 3.4628,
|
652 |
+
"num_input_tokens_seen": 175912960,
|
653 |
+
"step": 21500
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 31.0,
|
657 |
+
"eval_accuracy": 0.26106526631657917,
|
658 |
+
"eval_loss": 4.910233974456787,
|
659 |
+
"eval_runtime": 2.3245,
|
660 |
+
"eval_samples_per_second": 129.491,
|
661 |
+
"eval_steps_per_second": 16.348,
|
662 |
+
"num_input_tokens_seen": 179829760,
|
663 |
+
"step": 21979
|
664 |
+
},
|
665 |
+
{
|
666 |
+
"epoch": 31.029619181946405,
|
667 |
+
"grad_norm": 9.639541625976562,
|
668 |
+
"learning_rate": 1.8970380818053598e-05,
|
669 |
+
"loss": 3.4774,
|
670 |
+
"num_input_tokens_seen": 180001792,
|
671 |
+
"step": 22000
|
672 |
+
},
|
673 |
+
{
|
674 |
+
"epoch": 31.734837799717912,
|
675 |
+
"grad_norm": 8.425528526306152,
|
676 |
+
"learning_rate": 1.8265162200282088e-05,
|
677 |
+
"loss": 3.4478,
|
678 |
+
"num_input_tokens_seen": 184097792,
|
679 |
+
"step": 22500
|
680 |
+
},
|
681 |
+
{
|
682 |
+
"epoch": 32.0,
|
683 |
+
"eval_accuracy": 0.2584022629033882,
|
684 |
+
"eval_loss": 4.99893856048584,
|
685 |
+
"eval_runtime": 2.4209,
|
686 |
+
"eval_samples_per_second": 124.332,
|
687 |
+
"eval_steps_per_second": 15.696,
|
688 |
+
"num_input_tokens_seen": 185630720,
|
689 |
+
"step": 22688
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 32.44005641748942,
|
693 |
+
"grad_norm": 12.94858455657959,
|
694 |
+
"learning_rate": 1.755994358251058e-05,
|
695 |
+
"loss": 3.4539,
|
696 |
+
"num_input_tokens_seen": 188186624,
|
697 |
+
"step": 23000
|
698 |
+
},
|
699 |
+
{
|
700 |
+
"epoch": 33.0,
|
701 |
+
"eval_accuracy": 0.2561939186095225,
|
702 |
+
"eval_loss": 5.060764789581299,
|
703 |
+
"eval_runtime": 2.3544,
|
704 |
+
"eval_samples_per_second": 127.845,
|
705 |
+
"eval_steps_per_second": 16.14,
|
706 |
+
"num_input_tokens_seen": 191431680,
|
707 |
+
"step": 23397
|
708 |
+
},
|
709 |
+
{
|
710 |
+
"epoch": 33.14527503526093,
|
711 |
+
"grad_norm": 10.570552825927734,
|
712 |
+
"learning_rate": 1.685472496473907e-05,
|
713 |
+
"loss": 3.4532,
|
714 |
+
"num_input_tokens_seen": 192275456,
|
715 |
+
"step": 23500
|
716 |
+
},
|
717 |
+
{
|
718 |
+
"epoch": 33.85049365303244,
|
719 |
+
"grad_norm": 9.403538703918457,
|
720 |
+
"learning_rate": 1.614950634696756e-05,
|
721 |
+
"loss": 3.4486,
|
722 |
+
"num_input_tokens_seen": 196371456,
|
723 |
+
"step": 24000
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 34.0,
|
727 |
+
"eval_accuracy": 0.2573338139729738,
|
728 |
+
"eval_loss": 5.155101299285889,
|
729 |
+
"eval_runtime": 2.3548,
|
730 |
+
"eval_samples_per_second": 127.825,
|
731 |
+
"eval_steps_per_second": 16.137,
|
732 |
+
"num_input_tokens_seen": 197232640,
|
733 |
+
"step": 24106
|
734 |
+
},
|
735 |
+
{
|
736 |
+
"epoch": 34.55571227080395,
|
737 |
+
"grad_norm": 8.42175006866455,
|
738 |
+
"learning_rate": 1.5444287729196053e-05,
|
739 |
+
"loss": 3.4265,
|
740 |
+
"num_input_tokens_seen": 200460288,
|
741 |
+
"step": 24500
|
742 |
+
},
|
743 |
+
{
|
744 |
+
"epoch": 35.0,
|
745 |
+
"eval_accuracy": 0.2559276182682034,
|
746 |
+
"eval_loss": 5.19606876373291,
|
747 |
+
"eval_runtime": 2.3467,
|
748 |
+
"eval_samples_per_second": 128.266,
|
749 |
+
"eval_steps_per_second": 16.193,
|
750 |
+
"num_input_tokens_seen": 203033600,
|
751 |
+
"step": 24815
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 35.26093088857546,
|
755 |
+
"grad_norm": 9.095492362976074,
|
756 |
+
"learning_rate": 1.4739069111424541e-05,
|
757 |
+
"loss": 3.4376,
|
758 |
+
"num_input_tokens_seen": 204549120,
|
759 |
+
"step": 25000
|
760 |
+
},
|
761 |
+
{
|
762 |
+
"epoch": 35.966149506346966,
|
763 |
+
"grad_norm": 9.631875038146973,
|
764 |
+
"learning_rate": 1.4033850493653034e-05,
|
765 |
+
"loss": 3.4427,
|
766 |
+
"num_input_tokens_seen": 208645120,
|
767 |
+
"step": 25500
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 36.0,
|
771 |
+
"eval_accuracy": 0.2559828268755501,
|
772 |
+
"eval_loss": 5.287012100219727,
|
773 |
+
"eval_runtime": 2.3609,
|
774 |
+
"eval_samples_per_second": 127.496,
|
775 |
+
"eval_steps_per_second": 16.096,
|
776 |
+
"num_input_tokens_seen": 208834560,
|
777 |
+
"step": 25524
|
778 |
+
},
|
779 |
+
{
|
780 |
+
"epoch": 36.67136812411848,
|
781 |
+
"grad_norm": 9.127022743225098,
|
782 |
+
"learning_rate": 1.3328631875881522e-05,
|
783 |
+
"loss": 3.4203,
|
784 |
+
"num_input_tokens_seen": 212733952,
|
785 |
+
"step": 26000
|
786 |
+
},
|
787 |
+
{
|
788 |
+
"epoch": 37.0,
|
789 |
+
"eval_accuracy": 0.25666481555453796,
|
790 |
+
"eval_loss": 5.381103515625,
|
791 |
+
"eval_runtime": 2.408,
|
792 |
+
"eval_samples_per_second": 125.0,
|
793 |
+
"eval_steps_per_second": 15.781,
|
794 |
+
"num_input_tokens_seen": 214635520,
|
795 |
+
"step": 26233
|
796 |
+
},
|
797 |
+
{
|
798 |
+
"epoch": 37.37658674188999,
|
799 |
+
"grad_norm": 8.888460159301758,
|
800 |
+
"learning_rate": 1.2623413258110015e-05,
|
801 |
+
"loss": 3.4309,
|
802 |
+
"num_input_tokens_seen": 216822784,
|
803 |
+
"step": 26500
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 38.0,
|
807 |
+
"eval_accuracy": 0.2557100314039549,
|
808 |
+
"eval_loss": 5.395905017852783,
|
809 |
+
"eval_runtime": 2.3536,
|
810 |
+
"eval_samples_per_second": 127.891,
|
811 |
+
"eval_steps_per_second": 16.146,
|
812 |
+
"num_input_tokens_seen": 220436480,
|
813 |
+
"step": 26942
|
814 |
+
},
|
815 |
+
{
|
816 |
+
"epoch": 38.081805359661494,
|
817 |
+
"grad_norm": 9.691740036010742,
|
818 |
+
"learning_rate": 1.1918194640338505e-05,
|
819 |
+
"loss": 3.442,
|
820 |
+
"num_input_tokens_seen": 220911616,
|
821 |
+
"step": 27000
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 38.787023977433,
|
825 |
+
"grad_norm": 9.186783790588379,
|
826 |
+
"learning_rate": 1.1212976022566998e-05,
|
827 |
+
"loss": 3.4265,
|
828 |
+
"num_input_tokens_seen": 225007616,
|
829 |
+
"step": 27500
|
830 |
+
},
|
831 |
+
{
|
832 |
+
"epoch": 39.0,
|
833 |
+
"eval_accuracy": 0.25483643638182274,
|
834 |
+
"eval_loss": 5.532092571258545,
|
835 |
+
"eval_runtime": 2.3449,
|
836 |
+
"eval_samples_per_second": 128.361,
|
837 |
+
"eval_steps_per_second": 16.205,
|
838 |
+
"num_input_tokens_seen": 226237440,
|
839 |
+
"step": 27651
|
840 |
+
},
|
841 |
+
{
|
842 |
+
"epoch": 39.492242595204516,
|
843 |
+
"grad_norm": 8.784052848815918,
|
844 |
+
"learning_rate": 1.0507757404795487e-05,
|
845 |
+
"loss": 3.4267,
|
846 |
+
"num_input_tokens_seen": 229096448,
|
847 |
+
"step": 28000
|
848 |
+
},
|
849 |
+
{
|
850 |
+
"epoch": 40.0,
|
851 |
+
"eval_accuracy": 0.2528132033008252,
|
852 |
+
"eval_loss": 5.587483882904053,
|
853 |
+
"eval_runtime": 2.4783,
|
854 |
+
"eval_samples_per_second": 121.456,
|
855 |
+
"eval_steps_per_second": 15.333,
|
856 |
+
"num_input_tokens_seen": 232038400,
|
857 |
+
"step": 28360
|
858 |
+
},
|
859 |
+
{
|
860 |
+
"epoch": 40.19746121297602,
|
861 |
+
"grad_norm": 9.762134552001953,
|
862 |
+
"learning_rate": 9.802538787023979e-06,
|
863 |
+
"loss": 3.4331,
|
864 |
+
"num_input_tokens_seen": 233185280,
|
865 |
+
"step": 28500
|
866 |
+
},
|
867 |
+
{
|
868 |
+
"epoch": 40.90267983074753,
|
869 |
+
"grad_norm": 9.590001106262207,
|
870 |
+
"learning_rate": 9.097320169252468e-06,
|
871 |
+
"loss": 3.456,
|
872 |
+
"num_input_tokens_seen": 237281280,
|
873 |
+
"step": 29000
|
874 |
+
},
|
875 |
+
{
|
876 |
+
"epoch": 41.0,
|
877 |
+
"eval_accuracy": 0.25092312039048725,
|
878 |
+
"eval_loss": 5.615477561950684,
|
879 |
+
"eval_runtime": 2.4777,
|
880 |
+
"eval_samples_per_second": 121.485,
|
881 |
+
"eval_steps_per_second": 15.337,
|
882 |
+
"num_input_tokens_seen": 237839360,
|
883 |
+
"step": 29069
|
884 |
+
},
|
885 |
+
{
|
886 |
+
"epoch": 41.60789844851904,
|
887 |
+
"grad_norm": 9.302165985107422,
|
888 |
+
"learning_rate": 8.39210155148096e-06,
|
889 |
+
"loss": 3.445,
|
890 |
+
"num_input_tokens_seen": 241370112,
|
891 |
+
"step": 29500
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 42.0,
|
895 |
+
"eval_accuracy": 0.2526670628696135,
|
896 |
+
"eval_loss": 5.626107215881348,
|
897 |
+
"eval_runtime": 2.4409,
|
898 |
+
"eval_samples_per_second": 123.313,
|
899 |
+
"eval_steps_per_second": 15.568,
|
900 |
+
"num_input_tokens_seen": 243640320,
|
901 |
+
"step": 29778
|
902 |
+
},
|
903 |
+
{
|
904 |
+
"epoch": 42.31311706629055,
|
905 |
+
"grad_norm": 11.168075561523438,
|
906 |
+
"learning_rate": 7.686882933709451e-06,
|
907 |
+
"loss": 3.4674,
|
908 |
+
"num_input_tokens_seen": 245458944,
|
909 |
+
"step": 30000
|
910 |
+
},
|
911 |
+
{
|
912 |
+
"epoch": 43.0,
|
913 |
+
"eval_accuracy": 0.2473248182175414,
|
914 |
+
"eval_loss": 5.813519477844238,
|
915 |
+
"eval_runtime": 2.4619,
|
916 |
+
"eval_samples_per_second": 122.263,
|
917 |
+
"eval_steps_per_second": 15.435,
|
918 |
+
"num_input_tokens_seen": 249441280,
|
919 |
+
"step": 30487
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 43.01833568406206,
|
923 |
+
"grad_norm": 13.5941743850708,
|
924 |
+
"learning_rate": 6.9816643159379414e-06,
|
925 |
+
"loss": 3.4873,
|
926 |
+
"num_input_tokens_seen": 249547776,
|
927 |
+
"step": 30500
|
928 |
+
},
|
929 |
+
{
|
930 |
+
"epoch": 43.723554301833566,
|
931 |
+
"grad_norm": 10.031736373901367,
|
932 |
+
"learning_rate": 6.276445698166432e-06,
|
933 |
+
"loss": 3.4888,
|
934 |
+
"num_input_tokens_seen": 253643776,
|
935 |
+
"step": 31000
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 44.0,
|
939 |
+
"eval_accuracy": 0.24821465106536375,
|
940 |
+
"eval_loss": 5.875204086303711,
|
941 |
+
"eval_runtime": 2.3416,
|
942 |
+
"eval_samples_per_second": 128.546,
|
943 |
+
"eval_steps_per_second": 16.228,
|
944 |
+
"num_input_tokens_seen": 255242240,
|
945 |
+
"step": 31196
|
946 |
+
},
|
947 |
+
{
|
948 |
+
"epoch": 44.42877291960508,
|
949 |
+
"grad_norm": 11.021130561828613,
|
950 |
+
"learning_rate": 5.5712270803949224e-06,
|
951 |
+
"loss": 3.5004,
|
952 |
+
"num_input_tokens_seen": 257732608,
|
953 |
+
"step": 31500
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 45.0,
|
957 |
+
"eval_accuracy": 0.24905577043611551,
|
958 |
+
"eval_loss": 5.897055625915527,
|
959 |
+
"eval_runtime": 2.3591,
|
960 |
+
"eval_samples_per_second": 127.589,
|
961 |
+
"eval_steps_per_second": 16.108,
|
962 |
+
"num_input_tokens_seen": 261043200,
|
963 |
+
"step": 31905
|
964 |
+
},
|
965 |
+
{
|
966 |
+
"epoch": 45.13399153737659,
|
967 |
+
"grad_norm": 9.840156555175781,
|
968 |
+
"learning_rate": 4.866008462623414e-06,
|
969 |
+
"loss": 3.4993,
|
970 |
+
"num_input_tokens_seen": 261821440,
|
971 |
+
"step": 32000
|
972 |
+
},
|
973 |
+
{
|
974 |
+
"epoch": 45.839210155148095,
|
975 |
+
"grad_norm": 12.190547943115234,
|
976 |
+
"learning_rate": 4.160789844851904e-06,
|
977 |
+
"loss": 3.5048,
|
978 |
+
"num_input_tokens_seen": 265917440,
|
979 |
+
"step": 32500
|
980 |
+
},
|
981 |
+
{
|
982 |
+
"epoch": 46.0,
|
983 |
+
"eval_accuracy": 0.2463960145880626,
|
984 |
+
"eval_loss": 5.995114803314209,
|
985 |
+
"eval_runtime": 2.6148,
|
986 |
+
"eval_samples_per_second": 115.115,
|
987 |
+
"eval_steps_per_second": 14.533,
|
988 |
+
"num_input_tokens_seen": 266844160,
|
989 |
+
"step": 32614
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 46.5444287729196,
|
993 |
+
"grad_norm": 10.28553581237793,
|
994 |
+
"learning_rate": 3.455571227080395e-06,
|
995 |
+
"loss": 3.5265,
|
996 |
+
"num_input_tokens_seen": 270006272,
|
997 |
+
"step": 33000
|
998 |
+
},
|
999 |
+
{
|
1000 |
+
"epoch": 47.0,
|
1001 |
+
"eval_accuracy": 0.24732806578267943,
|
1002 |
+
"eval_loss": 5.983969688415527,
|
1003 |
+
"eval_runtime": 2.5908,
|
1004 |
+
"eval_samples_per_second": 116.181,
|
1005 |
+
"eval_steps_per_second": 14.667,
|
1006 |
+
"num_input_tokens_seen": 272645120,
|
1007 |
+
"step": 33323
|
1008 |
+
},
|
1009 |
+
{
|
1010 |
+
"epoch": 47.249647390691116,
|
1011 |
+
"grad_norm": 9.88039779663086,
|
1012 |
+
"learning_rate": 2.7503526093088857e-06,
|
1013 |
+
"loss": 3.5267,
|
1014 |
+
"num_input_tokens_seen": 274095104,
|
1015 |
+
"step": 33500
|
1016 |
+
},
|
1017 |
+
{
|
1018 |
+
"epoch": 47.95486600846262,
|
1019 |
+
"grad_norm": 18.002471923828125,
|
1020 |
+
"learning_rate": 2.0451339915373766e-06,
|
1021 |
+
"loss": 3.5454,
|
1022 |
+
"num_input_tokens_seen": 278191104,
|
1023 |
+
"step": 34000
|
1024 |
+
},
|
1025 |
+
{
|
1026 |
+
"epoch": 48.0,
|
1027 |
+
"eval_accuracy": 0.24613296181188155,
|
1028 |
+
"eval_loss": 6.077630043029785,
|
1029 |
+
"eval_runtime": 2.56,
|
1030 |
+
"eval_samples_per_second": 117.579,
|
1031 |
+
"eval_steps_per_second": 14.844,
|
1032 |
+
"num_input_tokens_seen": 278446080,
|
1033 |
+
"step": 34032
|
1034 |
+
},
|
1035 |
+
{
|
1036 |
+
"epoch": 48.66008462623413,
|
1037 |
+
"grad_norm": 9.459271430969238,
|
1038 |
+
"learning_rate": 1.3399153737658675e-06,
|
1039 |
+
"loss": 3.5405,
|
1040 |
+
"num_input_tokens_seen": 282279936,
|
1041 |
+
"step": 34500
|
1042 |
+
},
|
1043 |
+
{
|
1044 |
+
"epoch": 49.0,
|
1045 |
+
"eval_accuracy": 0.2454509731328936,
|
1046 |
+
"eval_loss": 6.075345993041992,
|
1047 |
+
"eval_runtime": 2.392,
|
1048 |
+
"eval_samples_per_second": 125.835,
|
1049 |
+
"eval_steps_per_second": 15.886,
|
1050 |
+
"num_input_tokens_seen": 284247040,
|
1051 |
+
"step": 34741
|
1052 |
+
},
|
1053 |
+
{
|
1054 |
+
"epoch": 49.365303244005645,
|
1055 |
+
"grad_norm": 10.473950386047363,
|
1056 |
+
"learning_rate": 6.346967559943583e-07,
|
1057 |
+
"loss": 3.5448,
|
1058 |
+
"num_input_tokens_seen": 286368768,
|
1059 |
+
"step": 35000
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 50.0,
|
1063 |
+
"eval_accuracy": 0.2452431289640592,
|
1064 |
+
"eval_loss": 6.103971004486084,
|
1065 |
+
"eval_runtime": 2.5249,
|
1066 |
+
"eval_samples_per_second": 119.214,
|
1067 |
+
"eval_steps_per_second": 15.05,
|
1068 |
+
"num_input_tokens_seen": 290048000,
|
1069 |
+
"step": 35450
|
1070 |
+
},
|
1071 |
+
{
|
1072 |
+
"epoch": 50.0,
|
1073 |
+
"num_input_tokens_seen": 290048000,
|
1074 |
+
"step": 35450,
|
1075 |
+
"total_flos": 7.7740363481088e+16,
|
1076 |
+
"train_loss": 3.9178111260633375,
|
1077 |
+
"train_runtime": 4788.9263,
|
1078 |
+
"train_samples_per_second": 59.147,
|
1079 |
+
"train_steps_per_second": 7.402,
|
1080 |
+
"train_tokens_per_second": 60566.395
|
1081 |
+
}
|
1082 |
+
],
|
1083 |
+
"logging_steps": 500,
|
1084 |
+
"max_steps": 35450,
|
1085 |
+
"num_input_tokens_seen": 290048000,
|
1086 |
+
"num_train_epochs": 50,
|
1087 |
+
"save_steps": 500,
|
1088 |
+
"stateful_callbacks": {
|
1089 |
+
"TrainerControl": {
|
1090 |
+
"args": {
|
1091 |
+
"should_epoch_stop": false,
|
1092 |
+
"should_evaluate": false,
|
1093 |
+
"should_log": false,
|
1094 |
+
"should_save": true,
|
1095 |
+
"should_training_stop": true
|
1096 |
+
},
|
1097 |
+
"attributes": {}
|
1098 |
+
}
|
1099 |
+
},
|
1100 |
+
"total_flos": 7.7740363481088e+16,
|
1101 |
+
"train_batch_size": 8,
|
1102 |
+
"trial_name": null,
|
1103 |
+
"trial_params": null
|
1104 |
+
}
|