Training in progress, epoch 36, checkpoint
Browse files- checkpoint-6750/adapter_config.json +1 -1
- checkpoint-6750/adapter_model.bin +3 -0
- checkpoint-6750/added_tokens.json +3 -3
- checkpoint-6750/optimizer.pt +2 -2
- checkpoint-6750/rng_state_0.pth +1 -1
- checkpoint-6750/rng_state_1.pth +1 -1
- checkpoint-6750/rng_state_2.pth +1 -1
- checkpoint-6750/rng_state_3.pth +1 -1
- checkpoint-6750/special_tokens_map.json +4 -12
- checkpoint-6750/tokenizer.json +0 -0
- checkpoint-6750/tokenizer.model +3 -0
- checkpoint-6750/tokenizer_config.json +17 -20
- checkpoint-6750/trainer_state.json +401 -482
- checkpoint-6750/training_args.bin +2 -2
checkpoint-6750/adapter_config.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"auto_mapping": null,
|
| 3 |
-
"base_model_name_or_path": "
|
| 4 |
"bias": "none",
|
| 5 |
"fan_in_fan_out": false,
|
| 6 |
"inference_mode": true,
|
|
|
|
| 1 |
{
|
| 2 |
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
|
| 4 |
"bias": "none",
|
| 5 |
"fan_in_fan_out": false,
|
| 6 |
"inference_mode": true,
|
checkpoint-6750/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:755c672a2a4bb2a96c1baaecb62c54b76cb96d62a788c4304deacd6e020884e6
|
| 3 |
+
size 143269386
|
checkpoint-6750/added_tokens.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"</s>": 2,
|
| 3 |
+
"<s>": 1,
|
| 4 |
+
"<unk>": 0
|
| 5 |
}
|
checkpoint-6750/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:29f4caacc1aa15bfaee4abf09f270486dc8b31a4237dd8d62a59a665761fc5c9
|
| 3 |
+
size 286585234
|
checkpoint-6750/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c7ee6a319f7884d4474c7fc78fdc3d72e9b1b5ff9494a4d07db7041e84a59bcc
|
| 3 |
size 15024
|
checkpoint-6750/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ef02f6715ea83ecea9ad0127edfa7590073f685c7ca3ba26771defcac4742b62
|
| 3 |
size 15024
|
checkpoint-6750/rng_state_2.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:40519a3e5213cab3c91fc4d5f397d0e14e311442fda26ad7e2a0b5e59e3b885d
|
| 3 |
size 15024
|
checkpoint-6750/rng_state_3.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:de5172dbe3c9ff95768519f9b7b1bbb1aee9087eb68c98bce2ad0909aec81656
|
| 3 |
size 15024
|
checkpoint-6750/special_tokens_map.json
CHANGED
|
@@ -1,14 +1,6 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
"eos_token": {
|
| 7 |
-
"content": "<|endoftext|>",
|
| 8 |
-
"lstrip": false,
|
| 9 |
-
"normalized": false,
|
| 10 |
-
"rstrip": false,
|
| 11 |
-
"single_word": false
|
| 12 |
-
},
|
| 13 |
-
"pad_token": "<|endoftext|>"
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"eos_token": "</s>",
|
| 4 |
+
"pad_token": "</s>",
|
| 5 |
+
"unk_token": "<unk>"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
}
|
checkpoint-6750/tokenizer.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-6750/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
checkpoint-6750/tokenizer_config.json
CHANGED
|
@@ -1,24 +1,23 @@
|
|
| 1 |
{
|
| 2 |
-
"add_prefix_space": false,
|
| 3 |
"added_tokens_decoder": {
|
| 4 |
-
"
|
| 5 |
-
"content": "
|
| 6 |
"lstrip": false,
|
| 7 |
"normalized": false,
|
| 8 |
"rstrip": false,
|
| 9 |
"single_word": false,
|
| 10 |
"special": true
|
| 11 |
},
|
| 12 |
-
"
|
| 13 |
-
"content": "
|
| 14 |
"lstrip": false,
|
| 15 |
"normalized": false,
|
| 16 |
"rstrip": false,
|
| 17 |
"single_word": false,
|
| 18 |
"special": true
|
| 19 |
},
|
| 20 |
-
"
|
| 21 |
-
"content": "
|
| 22 |
"lstrip": false,
|
| 23 |
"normalized": false,
|
| 24 |
"rstrip": false,
|
|
@@ -26,18 +25,16 @@
|
|
| 26 |
"special": true
|
| 27 |
}
|
| 28 |
},
|
| 29 |
-
"additional_special_tokens": [
|
| 30 |
-
|
| 31 |
-
"<|im_end|>"
|
| 32 |
-
],
|
| 33 |
-
"bos_token": null,
|
| 34 |
-
"chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 35 |
"clean_up_tokenization_spaces": false,
|
| 36 |
-
"eos_token": "
|
| 37 |
-
"
|
| 38 |
-
"model_max_length":
|
| 39 |
-
"pad_token": "
|
| 40 |
-
"
|
| 41 |
-
"
|
| 42 |
-
"
|
|
|
|
|
|
|
| 43 |
}
|
|
|
|
| 1 |
{
|
|
|
|
| 2 |
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<unk>",
|
| 5 |
"lstrip": false,
|
| 6 |
"normalized": false,
|
| 7 |
"rstrip": false,
|
| 8 |
"single_word": false,
|
| 9 |
"special": true
|
| 10 |
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<s>",
|
| 13 |
"lstrip": false,
|
| 14 |
"normalized": false,
|
| 15 |
"rstrip": false,
|
| 16 |
"single_word": false,
|
| 17 |
"special": true
|
| 18 |
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
"lstrip": false,
|
| 22 |
"normalized": false,
|
| 23 |
"rstrip": false,
|
|
|
|
| 25 |
"special": true
|
| 26 |
}
|
| 27 |
},
|
| 28 |
+
"additional_special_tokens": [],
|
| 29 |
+
"bos_token": "<s>",
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
"clean_up_tokenization_spaces": false,
|
| 31 |
+
"eos_token": "</s>",
|
| 32 |
+
"legacy": false,
|
| 33 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 34 |
+
"pad_token": "</s>",
|
| 35 |
+
"padding_side": "left",
|
| 36 |
+
"sp_model_kwargs": {},
|
| 37 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 38 |
+
"unk_token": "<unk>",
|
| 39 |
+
"use_default_system_prompt": true
|
| 40 |
}
|
checkpoint-6750/trainer_state.json
CHANGED
|
@@ -9,1034 +9,953 @@
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm": 0.5194531679153442,
|
| 14 |
"learning_rate": 0.0003,
|
| 15 |
-
"loss": 1.
|
| 16 |
"step": 100
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0
|
| 20 |
-
"eval_accuracy": 0.
|
| 21 |
-
"eval_loss": 1.
|
| 22 |
-
"eval_runtime":
|
| 23 |
-
"eval_samples_per_second":
|
| 24 |
-
"eval_steps_per_second":
|
| 25 |
"step": 187
|
| 26 |
},
|
| 27 |
{
|
| 28 |
-
"epoch": 0
|
| 29 |
-
"eval_exact_match":
|
| 30 |
-
"eval_f1":
|
| 31 |
"step": 187
|
| 32 |
},
|
| 33 |
{
|
| 34 |
-
"epoch": 1.
|
| 35 |
-
"grad_norm": 0.6734262704849243,
|
| 36 |
"learning_rate": 0.0003,
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 200
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.6,
|
| 42 |
-
"grad_norm": 0.7422465682029724,
|
| 43 |
"learning_rate": 0.0003,
|
| 44 |
-
"loss": 1.
|
| 45 |
"step": 300
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 2.0,
|
| 49 |
-
"eval_accuracy": 0.
|
| 50 |
-
"eval_loss": 1.
|
| 51 |
-
"eval_runtime":
|
| 52 |
-
"eval_samples_per_second":
|
| 53 |
-
"eval_steps_per_second":
|
| 54 |
"step": 375
|
| 55 |
},
|
| 56 |
{
|
| 57 |
"epoch": 2.0,
|
| 58 |
-
"eval_exact_match":
|
| 59 |
-
"eval_f1":
|
| 60 |
"step": 375
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"epoch": 2.
|
| 64 |
-
"grad_norm": 0.828364908695221,
|
| 65 |
"learning_rate": 0.0003,
|
| 66 |
-
"loss":
|
| 67 |
"step": 400
|
| 68 |
},
|
| 69 |
{
|
| 70 |
-
"epoch": 2.
|
| 71 |
-
"grad_norm": 0.9025559425354004,
|
| 72 |
"learning_rate": 0.0003,
|
| 73 |
-
"loss": 0.
|
| 74 |
"step": 500
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch":
|
| 78 |
-
"eval_accuracy": 0.
|
| 79 |
-
"eval_loss": 1.
|
| 80 |
-
"eval_runtime":
|
| 81 |
-
"eval_samples_per_second":
|
| 82 |
-
"eval_steps_per_second":
|
| 83 |
"step": 562
|
| 84 |
},
|
| 85 |
{
|
| 86 |
-
"epoch":
|
| 87 |
-
"eval_exact_match":
|
| 88 |
-
"eval_f1":
|
| 89 |
"step": 562
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"epoch": 3.2,
|
| 93 |
-
"grad_norm": 0.8217537999153137,
|
| 94 |
"learning_rate": 0.0003,
|
| 95 |
-
"loss": 0.
|
| 96 |
"step": 600
|
| 97 |
},
|
| 98 |
{
|
| 99 |
-
"epoch": 3.
|
| 100 |
-
"grad_norm": 0.8315110206604004,
|
| 101 |
"learning_rate": 0.0003,
|
| 102 |
-
"loss": 0.
|
| 103 |
"step": 700
|
| 104 |
},
|
| 105 |
{
|
| 106 |
"epoch": 4.0,
|
| 107 |
-
"eval_accuracy": 0.
|
| 108 |
-
"eval_loss": 1.
|
| 109 |
-
"eval_runtime":
|
| 110 |
-
"eval_samples_per_second":
|
| 111 |
-
"eval_steps_per_second":
|
| 112 |
"step": 750
|
| 113 |
},
|
| 114 |
{
|
| 115 |
"epoch": 4.0,
|
| 116 |
-
"eval_exact_match":
|
| 117 |
-
"eval_f1":
|
| 118 |
"step": 750
|
| 119 |
},
|
| 120 |
{
|
| 121 |
-
"epoch": 4.
|
| 122 |
-
"grad_norm": 0.6763857007026672,
|
| 123 |
"learning_rate": 0.0003,
|
| 124 |
-
"loss": 0.
|
| 125 |
"step": 800
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 4.8,
|
| 129 |
-
"grad_norm": 0.8408166766166687,
|
| 130 |
"learning_rate": 0.0003,
|
| 131 |
-
"loss": 0.
|
| 132 |
"step": 900
|
| 133 |
},
|
| 134 |
{
|
| 135 |
-
"epoch":
|
| 136 |
-
"eval_accuracy": 0.
|
| 137 |
-
"eval_loss": 1.
|
| 138 |
-
"eval_runtime":
|
| 139 |
-
"eval_samples_per_second":
|
| 140 |
-
"eval_steps_per_second":
|
| 141 |
"step": 937
|
| 142 |
},
|
| 143 |
{
|
| 144 |
-
"epoch":
|
| 145 |
-
"eval_exact_match":
|
| 146 |
-
"eval_f1":
|
| 147 |
"step": 937
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch": 5.
|
| 151 |
-
"grad_norm": 0.6721271276473999,
|
| 152 |
"learning_rate": 0.0003,
|
| 153 |
-
"loss": 0.
|
| 154 |
"step": 1000
|
| 155 |
},
|
| 156 |
{
|
| 157 |
-
"epoch": 5.
|
| 158 |
-
"grad_norm": 0.9258321523666382,
|
| 159 |
"learning_rate": 0.0003,
|
| 160 |
-
"loss": 0.
|
| 161 |
"step": 1100
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 6.0,
|
| 165 |
-
"eval_accuracy": 0.
|
| 166 |
-
"eval_loss":
|
| 167 |
-
"eval_runtime":
|
| 168 |
-
"eval_samples_per_second":
|
| 169 |
-
"eval_steps_per_second":
|
| 170 |
"step": 1125
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 6.0,
|
| 174 |
-
"eval_exact_match":
|
| 175 |
-
"eval_f1":
|
| 176 |
"step": 1125
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 6.4,
|
| 180 |
-
"grad_norm": 0.6029295921325684,
|
| 181 |
"learning_rate": 0.0003,
|
| 182 |
-
"loss": 0.
|
| 183 |
"step": 1200
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"epoch": 6.
|
| 187 |
-
"grad_norm": 0.5719736814498901,
|
| 188 |
"learning_rate": 0.0003,
|
| 189 |
-
"loss": 0.
|
| 190 |
"step": 1300
|
| 191 |
},
|
| 192 |
{
|
| 193 |
-
"epoch":
|
| 194 |
-
"eval_accuracy": 0.
|
| 195 |
-
"eval_loss":
|
| 196 |
-
"eval_runtime":
|
| 197 |
-
"eval_samples_per_second":
|
| 198 |
-
"eval_steps_per_second":
|
| 199 |
"step": 1312
|
| 200 |
},
|
| 201 |
{
|
| 202 |
-
"epoch":
|
| 203 |
-
"eval_exact_match":
|
| 204 |
-
"eval_f1":
|
| 205 |
"step": 1312
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch": 7.
|
| 209 |
-
"grad_norm": 0.49924588203430176,
|
| 210 |
"learning_rate": 0.0003,
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 1400
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 8.0,
|
| 216 |
-
"grad_norm": 0.5360875129699707,
|
| 217 |
"learning_rate": 0.0003,
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 1500
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 8.0,
|
| 223 |
-
"eval_accuracy": 0.
|
| 224 |
-
"eval_loss":
|
| 225 |
-
"eval_runtime":
|
| 226 |
-
"eval_samples_per_second":
|
| 227 |
-
"eval_steps_per_second":
|
| 228 |
"step": 1500
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 8.0,
|
| 232 |
-
"eval_exact_match":
|
| 233 |
-
"eval_f1":
|
| 234 |
"step": 1500
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch": 8.
|
| 238 |
-
"grad_norm": 0.44560280442237854,
|
| 239 |
"learning_rate": 0.0003,
|
| 240 |
-
"loss": 0.
|
| 241 |
"step": 1600
|
| 242 |
},
|
| 243 |
{
|
| 244 |
-
"epoch":
|
| 245 |
-
"eval_accuracy": 0.
|
| 246 |
-
"eval_loss":
|
| 247 |
-
"eval_runtime":
|
| 248 |
-
"eval_samples_per_second":
|
| 249 |
-
"eval_steps_per_second":
|
| 250 |
"step": 1687
|
| 251 |
},
|
| 252 |
{
|
| 253 |
-
"epoch":
|
| 254 |
-
"eval_exact_match":
|
| 255 |
-
"eval_f1":
|
| 256 |
"step": 1687
|
| 257 |
},
|
| 258 |
{
|
| 259 |
-
"epoch": 9.
|
| 260 |
-
"grad_norm": 0.42847225069999695,
|
| 261 |
"learning_rate": 0.0003,
|
| 262 |
-
"loss": 0.
|
| 263 |
"step": 1700
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 9.6,
|
| 267 |
-
"grad_norm": 0.4186452627182007,
|
| 268 |
"learning_rate": 0.0003,
|
| 269 |
-
"loss": 0.
|
| 270 |
"step": 1800
|
| 271 |
},
|
| 272 |
{
|
| 273 |
"epoch": 10.0,
|
| 274 |
-
"eval_accuracy": 0.
|
| 275 |
-
"eval_loss":
|
| 276 |
-
"eval_runtime":
|
| 277 |
-
"eval_samples_per_second":
|
| 278 |
-
"eval_steps_per_second":
|
| 279 |
"step": 1875
|
| 280 |
},
|
| 281 |
{
|
| 282 |
"epoch": 10.0,
|
| 283 |
-
"eval_exact_match":
|
| 284 |
-
"eval_f1":
|
| 285 |
"step": 1875
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"epoch": 10.
|
| 289 |
-
"grad_norm": 0.427721232175827,
|
| 290 |
"learning_rate": 0.0003,
|
| 291 |
-
"loss": 0.
|
| 292 |
"step": 1900
|
| 293 |
},
|
| 294 |
{
|
| 295 |
-
"epoch": 10.
|
| 296 |
-
"grad_norm": 0.4298701584339142,
|
| 297 |
"learning_rate": 0.0003,
|
| 298 |
-
"loss": 0.
|
| 299 |
"step": 2000
|
| 300 |
},
|
| 301 |
{
|
| 302 |
-
"epoch":
|
| 303 |
-
"eval_accuracy": 0.
|
| 304 |
-
"eval_loss":
|
| 305 |
-
"eval_runtime":
|
| 306 |
-
"eval_samples_per_second":
|
| 307 |
-
"eval_steps_per_second":
|
| 308 |
"step": 2062
|
| 309 |
},
|
| 310 |
{
|
| 311 |
-
"epoch":
|
| 312 |
-
"eval_exact_match":
|
| 313 |
-
"eval_f1":
|
| 314 |
"step": 2062
|
| 315 |
},
|
| 316 |
{
|
| 317 |
"epoch": 11.2,
|
| 318 |
-
"grad_norm": 0.40195176005363464,
|
| 319 |
"learning_rate": 0.0003,
|
| 320 |
-
"loss": 0.
|
| 321 |
"step": 2100
|
| 322 |
},
|
| 323 |
{
|
| 324 |
-
"epoch": 11.
|
| 325 |
-
"grad_norm": 0.6030442118644714,
|
| 326 |
"learning_rate": 0.0003,
|
| 327 |
-
"loss": 0.
|
| 328 |
"step": 2200
|
| 329 |
},
|
| 330 |
{
|
| 331 |
"epoch": 12.0,
|
| 332 |
-
"eval_accuracy": 0.
|
| 333 |
-
"eval_loss":
|
| 334 |
-
"eval_runtime":
|
| 335 |
-
"eval_samples_per_second":
|
| 336 |
-
"eval_steps_per_second":
|
| 337 |
"step": 2250
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 12.0,
|
| 341 |
-
"eval_exact_match":
|
| 342 |
-
"eval_f1":
|
| 343 |
"step": 2250
|
| 344 |
},
|
| 345 |
{
|
| 346 |
-
"epoch": 12.
|
| 347 |
-
"grad_norm": 0.4596252143383026,
|
| 348 |
"learning_rate": 0.0003,
|
| 349 |
-
"loss": 0.
|
| 350 |
"step": 2300
|
| 351 |
},
|
| 352 |
{
|
| 353 |
"epoch": 12.8,
|
| 354 |
-
"grad_norm": 0.4469389021396637,
|
| 355 |
"learning_rate": 0.0003,
|
| 356 |
-
"loss": 0.
|
| 357 |
"step": 2400
|
| 358 |
},
|
| 359 |
{
|
| 360 |
-
"epoch":
|
| 361 |
-
"eval_accuracy": 0.
|
| 362 |
-
"eval_loss":
|
| 363 |
-
"eval_runtime":
|
| 364 |
-
"eval_samples_per_second":
|
| 365 |
-
"eval_steps_per_second":
|
| 366 |
"step": 2437
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch":
|
| 370 |
-
"eval_exact_match":
|
| 371 |
-
"eval_f1":
|
| 372 |
"step": 2437
|
| 373 |
},
|
| 374 |
{
|
| 375 |
-
"epoch": 13.
|
| 376 |
-
"grad_norm": 0.4875074028968811,
|
| 377 |
"learning_rate": 0.0003,
|
| 378 |
-
"loss": 0.
|
| 379 |
"step": 2500
|
| 380 |
},
|
| 381 |
{
|
| 382 |
-
"epoch": 13.
|
| 383 |
-
"grad_norm": 0.5969555974006653,
|
| 384 |
"learning_rate": 0.0003,
|
| 385 |
-
"loss": 0.
|
| 386 |
"step": 2600
|
| 387 |
},
|
| 388 |
{
|
| 389 |
"epoch": 14.0,
|
| 390 |
-
"eval_accuracy": 0.
|
| 391 |
-
"eval_loss":
|
| 392 |
-
"eval_runtime":
|
| 393 |
-
"eval_samples_per_second":
|
| 394 |
-
"eval_steps_per_second":
|
| 395 |
"step": 2625
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 14.0,
|
| 399 |
-
"eval_exact_match":
|
| 400 |
-
"eval_f1":
|
| 401 |
"step": 2625
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 14.4,
|
| 405 |
-
"grad_norm": 0.3906678259372711,
|
| 406 |
"learning_rate": 0.0003,
|
| 407 |
-
"loss": 0.
|
| 408 |
"step": 2700
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch": 14.
|
| 412 |
-
"grad_norm": 0.6833591461181641,
|
| 413 |
"learning_rate": 0.0003,
|
| 414 |
-
"loss": 0.
|
| 415 |
"step": 2800
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch":
|
| 419 |
-
"eval_accuracy": 0.
|
| 420 |
-
"eval_loss":
|
| 421 |
-
"eval_runtime":
|
| 422 |
-
"eval_samples_per_second":
|
| 423 |
-
"eval_steps_per_second":
|
| 424 |
"step": 2812
|
| 425 |
},
|
| 426 |
{
|
| 427 |
-
"epoch":
|
| 428 |
-
"eval_exact_match":
|
| 429 |
-
"eval_f1":
|
| 430 |
"step": 2812
|
| 431 |
},
|
| 432 |
{
|
| 433 |
-
"epoch": 15.
|
| 434 |
-
"grad_norm": 0.6206196546554565,
|
| 435 |
"learning_rate": 0.0003,
|
| 436 |
-
"loss": 0.
|
| 437 |
"step": 2900
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 16.0,
|
| 441 |
-
"grad_norm": 0.4821285903453827,
|
| 442 |
"learning_rate": 0.0003,
|
| 443 |
-
"loss": 0.
|
| 444 |
"step": 3000
|
| 445 |
},
|
| 446 |
{
|
| 447 |
"epoch": 16.0,
|
| 448 |
-
"eval_accuracy": 0.
|
| 449 |
-
"eval_loss":
|
| 450 |
-
"eval_runtime":
|
| 451 |
-
"eval_samples_per_second":
|
| 452 |
-
"eval_steps_per_second":
|
| 453 |
"step": 3000
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 16.0,
|
| 457 |
-
"eval_exact_match":
|
| 458 |
-
"eval_f1":
|
| 459 |
"step": 3000
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch": 16.
|
| 463 |
-
"grad_norm": 0.7102397084236145,
|
| 464 |
"learning_rate": 0.0003,
|
| 465 |
-
"loss": 0.
|
| 466 |
"step": 3100
|
| 467 |
},
|
| 468 |
{
|
| 469 |
-
"epoch":
|
| 470 |
-
"eval_accuracy": 0.
|
| 471 |
-
"eval_loss":
|
| 472 |
-
"eval_runtime":
|
| 473 |
-
"eval_samples_per_second":
|
| 474 |
-
"eval_steps_per_second":
|
| 475 |
"step": 3187
|
| 476 |
},
|
| 477 |
{
|
| 478 |
-
"epoch":
|
| 479 |
-
"eval_exact_match":
|
| 480 |
-
"eval_f1":
|
| 481 |
"step": 3187
|
| 482 |
},
|
| 483 |
{
|
| 484 |
-
"epoch": 17.
|
| 485 |
-
"grad_norm": 0.4603720009326935,
|
| 486 |
"learning_rate": 0.0003,
|
| 487 |
-
"loss": 0.
|
| 488 |
"step": 3200
|
| 489 |
},
|
| 490 |
{
|
| 491 |
"epoch": 17.6,
|
| 492 |
-
"grad_norm": 0.45880746841430664,
|
| 493 |
"learning_rate": 0.0003,
|
| 494 |
-
"loss": 0.
|
| 495 |
"step": 3300
|
| 496 |
},
|
| 497 |
{
|
| 498 |
"epoch": 18.0,
|
| 499 |
-
"eval_accuracy": 0.
|
| 500 |
-
"eval_loss":
|
| 501 |
-
"eval_runtime":
|
| 502 |
-
"eval_samples_per_second":
|
| 503 |
-
"eval_steps_per_second":
|
| 504 |
"step": 3375
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 18.0,
|
| 508 |
-
"eval_exact_match":
|
| 509 |
-
"eval_f1":
|
| 510 |
"step": 3375
|
| 511 |
},
|
| 512 |
{
|
| 513 |
-
"epoch": 18.
|
| 514 |
-
"grad_norm": 0.4228228032588959,
|
| 515 |
"learning_rate": 0.0003,
|
| 516 |
-
"loss": 0.
|
| 517 |
"step": 3400
|
| 518 |
},
|
| 519 |
{
|
| 520 |
-
"epoch": 18.
|
| 521 |
-
"grad_norm": 0.42700931429862976,
|
| 522 |
"learning_rate": 0.0003,
|
| 523 |
-
"loss": 0.
|
| 524 |
"step": 3500
|
| 525 |
},
|
| 526 |
{
|
| 527 |
-
"epoch":
|
| 528 |
-
"eval_accuracy": 0.
|
| 529 |
-
"eval_loss": 2.
|
| 530 |
-
"eval_runtime":
|
| 531 |
-
"eval_samples_per_second":
|
| 532 |
-
"eval_steps_per_second":
|
| 533 |
"step": 3562
|
| 534 |
},
|
| 535 |
{
|
| 536 |
-
"epoch":
|
| 537 |
-
"eval_exact_match":
|
| 538 |
-
"eval_f1":
|
| 539 |
"step": 3562
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 19.2,
|
| 543 |
-
"grad_norm": 0.41856566071510315,
|
| 544 |
"learning_rate": 0.0003,
|
| 545 |
-
"loss":
|
| 546 |
"step": 3600
|
| 547 |
},
|
| 548 |
{
|
| 549 |
-
"epoch": 19.
|
| 550 |
-
"grad_norm": 0.4635602831840515,
|
| 551 |
"learning_rate": 0.0003,
|
| 552 |
-
"loss": 0.
|
| 553 |
"step": 3700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 20.0,
|
| 557 |
-
"eval_accuracy": 0.
|
| 558 |
-
"eval_loss":
|
| 559 |
-
"eval_runtime":
|
| 560 |
-
"eval_samples_per_second":
|
| 561 |
-
"eval_steps_per_second":
|
| 562 |
"step": 3750
|
| 563 |
},
|
| 564 |
{
|
| 565 |
"epoch": 20.0,
|
| 566 |
-
"eval_exact_match":
|
| 567 |
-
"eval_f1":
|
| 568 |
"step": 3750
|
| 569 |
},
|
| 570 |
{
|
| 571 |
-
"epoch": 20.
|
| 572 |
-
"grad_norm": 0.3601387143135071,
|
| 573 |
"learning_rate": 0.0003,
|
| 574 |
-
"loss": 0.
|
| 575 |
"step": 3800
|
| 576 |
},
|
| 577 |
{
|
| 578 |
"epoch": 20.8,
|
| 579 |
-
"grad_norm": 0.6374465227127075,
|
| 580 |
"learning_rate": 0.0003,
|
| 581 |
-
"loss": 0.
|
| 582 |
"step": 3900
|
| 583 |
},
|
| 584 |
{
|
| 585 |
-
"epoch":
|
| 586 |
-
"eval_accuracy": 0.
|
| 587 |
-
"eval_loss":
|
| 588 |
-
"eval_runtime":
|
| 589 |
-
"eval_samples_per_second":
|
| 590 |
-
"eval_steps_per_second":
|
| 591 |
"step": 3937
|
| 592 |
},
|
| 593 |
{
|
| 594 |
-
"epoch":
|
| 595 |
-
"eval_exact_match":
|
| 596 |
-
"eval_f1":
|
| 597 |
"step": 3937
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch": 21.
|
| 601 |
-
"grad_norm": 0.7570663690567017,
|
| 602 |
"learning_rate": 0.0003,
|
| 603 |
-
"loss": 0.
|
| 604 |
"step": 4000
|
| 605 |
},
|
| 606 |
{
|
| 607 |
-
"epoch": 21.
|
| 608 |
-
"grad_norm": 0.8324512839317322,
|
| 609 |
"learning_rate": 0.0003,
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 4100
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 22.0,
|
| 615 |
-
"eval_accuracy": 0.
|
| 616 |
-
"eval_loss":
|
| 617 |
-
"eval_runtime":
|
| 618 |
-
"eval_samples_per_second":
|
| 619 |
-
"eval_steps_per_second":
|
| 620 |
"step": 4125
|
| 621 |
},
|
| 622 |
{
|
| 623 |
"epoch": 22.0,
|
| 624 |
-
"eval_exact_match":
|
| 625 |
-
"eval_f1":
|
| 626 |
"step": 4125
|
| 627 |
},
|
| 628 |
{
|
| 629 |
"epoch": 22.4,
|
| 630 |
-
"grad_norm": 0.40705105662345886,
|
| 631 |
"learning_rate": 0.0003,
|
| 632 |
-
"loss":
|
| 633 |
"step": 4200
|
| 634 |
},
|
| 635 |
{
|
| 636 |
-
"epoch": 22.
|
| 637 |
-
"grad_norm": 0.38530829548835754,
|
| 638 |
"learning_rate": 0.0003,
|
| 639 |
-
"loss": 0.
|
| 640 |
"step": 4300
|
| 641 |
},
|
| 642 |
{
|
| 643 |
-
"epoch":
|
| 644 |
-
"eval_accuracy": 0.
|
| 645 |
-
"eval_loss":
|
| 646 |
-
"eval_runtime":
|
| 647 |
-
"eval_samples_per_second":
|
| 648 |
-
"eval_steps_per_second":
|
| 649 |
"step": 4312
|
| 650 |
},
|
| 651 |
{
|
| 652 |
-
"epoch":
|
| 653 |
-
"eval_exact_match":
|
| 654 |
-
"eval_f1":
|
| 655 |
"step": 4312
|
| 656 |
},
|
| 657 |
{
|
| 658 |
-
"epoch": 23.
|
| 659 |
-
"grad_norm": 0.4574826955795288,
|
| 660 |
"learning_rate": 0.0003,
|
| 661 |
-
"loss": 0.
|
| 662 |
"step": 4400
|
| 663 |
},
|
| 664 |
{
|
| 665 |
"epoch": 24.0,
|
| 666 |
-
"grad_norm": 0.5842512249946594,
|
| 667 |
"learning_rate": 0.0003,
|
| 668 |
-
"loss": 0.
|
| 669 |
"step": 4500
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 24.0,
|
| 673 |
-
"eval_accuracy": 0.
|
| 674 |
-
"eval_loss":
|
| 675 |
-
"eval_runtime":
|
| 676 |
-
"eval_samples_per_second":
|
| 677 |
-
"eval_steps_per_second":
|
| 678 |
"step": 4500
|
| 679 |
},
|
| 680 |
{
|
| 681 |
"epoch": 24.0,
|
| 682 |
-
"eval_exact_match":
|
| 683 |
-
"eval_f1":
|
| 684 |
"step": 4500
|
| 685 |
},
|
| 686 |
{
|
| 687 |
-
"epoch": 24.
|
| 688 |
-
"grad_norm": 0.41557422280311584,
|
| 689 |
"learning_rate": 0.0003,
|
| 690 |
-
"loss": 0.
|
| 691 |
"step": 4600
|
| 692 |
},
|
| 693 |
{
|
| 694 |
-
"epoch":
|
| 695 |
-
"eval_accuracy": 0.
|
| 696 |
-
"eval_loss":
|
| 697 |
-
"eval_runtime":
|
| 698 |
-
"eval_samples_per_second":
|
| 699 |
-
"eval_steps_per_second":
|
| 700 |
"step": 4687
|
| 701 |
},
|
| 702 |
{
|
| 703 |
-
"epoch":
|
| 704 |
-
"eval_exact_match":
|
| 705 |
-
"eval_f1":
|
| 706 |
"step": 4687
|
| 707 |
},
|
| 708 |
{
|
| 709 |
-
"epoch": 25.
|
| 710 |
-
"grad_norm": 0.46733421087265015,
|
| 711 |
"learning_rate": 0.0003,
|
| 712 |
-
"loss": 0.
|
| 713 |
"step": 4700
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 25.6,
|
| 717 |
-
"grad_norm": 0.3511659502983093,
|
| 718 |
"learning_rate": 0.0003,
|
| 719 |
-
"loss": 0.
|
| 720 |
"step": 4800
|
| 721 |
},
|
| 722 |
{
|
| 723 |
"epoch": 26.0,
|
| 724 |
-
"eval_accuracy": 0.
|
| 725 |
-
"eval_loss":
|
| 726 |
-
"eval_runtime":
|
| 727 |
-
"eval_samples_per_second":
|
| 728 |
-
"eval_steps_per_second":
|
| 729 |
"step": 4875
|
| 730 |
},
|
| 731 |
{
|
| 732 |
"epoch": 26.0,
|
| 733 |
-
"eval_exact_match":
|
| 734 |
-
"eval_f1":
|
| 735 |
"step": 4875
|
| 736 |
},
|
| 737 |
{
|
| 738 |
-
"epoch": 26.
|
| 739 |
-
"grad_norm": 0.7332673072814941,
|
| 740 |
"learning_rate": 0.0003,
|
| 741 |
-
"loss": 0.
|
| 742 |
"step": 4900
|
| 743 |
},
|
| 744 |
{
|
| 745 |
-
"epoch": 26.
|
| 746 |
-
"grad_norm": 0.49349483847618103,
|
| 747 |
"learning_rate": 0.0003,
|
| 748 |
-
"loss": 0.
|
| 749 |
"step": 5000
|
| 750 |
},
|
| 751 |
{
|
| 752 |
-
"epoch":
|
| 753 |
-
"eval_accuracy": 0.
|
| 754 |
-
"eval_loss":
|
| 755 |
-
"eval_runtime":
|
| 756 |
-
"eval_samples_per_second":
|
| 757 |
-
"eval_steps_per_second":
|
| 758 |
"step": 5062
|
| 759 |
},
|
| 760 |
{
|
| 761 |
-
"epoch":
|
| 762 |
-
"eval_exact_match":
|
| 763 |
-
"eval_f1":
|
| 764 |
"step": 5062
|
| 765 |
},
|
| 766 |
{
|
| 767 |
"epoch": 27.2,
|
| 768 |
-
"grad_norm": 0.5462525486946106,
|
| 769 |
"learning_rate": 0.0003,
|
| 770 |
-
"loss": 0.
|
| 771 |
"step": 5100
|
| 772 |
},
|
| 773 |
{
|
| 774 |
-
"epoch": 27.
|
| 775 |
-
"grad_norm": 0.3605049252510071,
|
| 776 |
"learning_rate": 0.0003,
|
| 777 |
-
"loss": 0.
|
| 778 |
"step": 5200
|
| 779 |
},
|
| 780 |
{
|
| 781 |
"epoch": 28.0,
|
| 782 |
-
"eval_accuracy": 0.
|
| 783 |
-
"eval_loss":
|
| 784 |
-
"eval_runtime":
|
| 785 |
-
"eval_samples_per_second":
|
| 786 |
-
"eval_steps_per_second":
|
| 787 |
"step": 5250
|
| 788 |
},
|
| 789 |
{
|
| 790 |
"epoch": 28.0,
|
| 791 |
-
"eval_exact_match":
|
| 792 |
-
"eval_f1":
|
| 793 |
"step": 5250
|
| 794 |
},
|
| 795 |
{
|
| 796 |
-
"epoch": 28.
|
| 797 |
-
"grad_norm": 0.9747931361198425,
|
| 798 |
"learning_rate": 0.0003,
|
| 799 |
-
"loss": 0.
|
| 800 |
"step": 5300
|
| 801 |
},
|
| 802 |
{
|
| 803 |
"epoch": 28.8,
|
| 804 |
-
"grad_norm": 0.45107710361480713,
|
| 805 |
"learning_rate": 0.0003,
|
| 806 |
-
"loss": 0.
|
| 807 |
"step": 5400
|
| 808 |
},
|
| 809 |
{
|
| 810 |
-
"epoch":
|
| 811 |
-
"eval_accuracy": 0.
|
| 812 |
-
"eval_loss":
|
| 813 |
-
"eval_runtime": 5.
|
| 814 |
-
"eval_samples_per_second":
|
| 815 |
-
"eval_steps_per_second":
|
| 816 |
"step": 5437
|
| 817 |
},
|
| 818 |
{
|
| 819 |
-
"epoch":
|
| 820 |
-
"eval_exact_match":
|
| 821 |
-
"eval_f1":
|
| 822 |
"step": 5437
|
| 823 |
},
|
| 824 |
{
|
| 825 |
-
"epoch": 29.
|
| 826 |
-
"grad_norm": 0.6405345797538757,
|
| 827 |
"learning_rate": 0.0003,
|
| 828 |
-
"loss": 0.
|
| 829 |
"step": 5500
|
| 830 |
},
|
| 831 |
{
|
| 832 |
-
"epoch": 29.
|
| 833 |
-
"grad_norm": 0.7235819697380066,
|
| 834 |
"learning_rate": 0.0003,
|
| 835 |
-
"loss": 0.
|
| 836 |
"step": 5600
|
| 837 |
},
|
| 838 |
{
|
| 839 |
"epoch": 30.0,
|
| 840 |
-
"eval_accuracy": 0.
|
| 841 |
-
"eval_loss":
|
| 842 |
-
"eval_runtime":
|
| 843 |
-
"eval_samples_per_second":
|
| 844 |
-
"eval_steps_per_second":
|
| 845 |
"step": 5625
|
| 846 |
},
|
| 847 |
{
|
| 848 |
"epoch": 30.0,
|
| 849 |
-
"eval_exact_match":
|
| 850 |
-
"eval_f1":
|
| 851 |
"step": 5625
|
| 852 |
},
|
| 853 |
{
|
| 854 |
"epoch": 30.4,
|
| 855 |
-
"grad_norm": 0.3755841851234436,
|
| 856 |
"learning_rate": 0.0003,
|
| 857 |
-
"loss": 0.
|
| 858 |
"step": 5700
|
| 859 |
},
|
| 860 |
{
|
| 861 |
-
"epoch": 30.
|
| 862 |
-
"grad_norm": 0.36230114102363586,
|
| 863 |
"learning_rate": 0.0003,
|
| 864 |
-
"loss": 0.
|
| 865 |
"step": 5800
|
| 866 |
},
|
| 867 |
{
|
| 868 |
-
"epoch":
|
| 869 |
-
"eval_accuracy": 0.
|
| 870 |
-
"eval_loss":
|
| 871 |
-
"eval_runtime":
|
| 872 |
-
"eval_samples_per_second":
|
| 873 |
-
"eval_steps_per_second":
|
| 874 |
"step": 5812
|
| 875 |
},
|
| 876 |
{
|
| 877 |
-
"epoch":
|
| 878 |
-
"eval_exact_match":
|
| 879 |
-
"eval_f1":
|
| 880 |
"step": 5812
|
| 881 |
},
|
| 882 |
{
|
| 883 |
-
"epoch": 31.
|
| 884 |
-
"grad_norm": 0.5876615047454834,
|
| 885 |
"learning_rate": 0.0003,
|
| 886 |
-
"loss": 0.
|
| 887 |
"step": 5900
|
| 888 |
},
|
| 889 |
{
|
| 890 |
"epoch": 32.0,
|
| 891 |
-
"grad_norm": 0.8609116673469543,
|
| 892 |
"learning_rate": 0.0003,
|
| 893 |
-
"loss": 0.
|
| 894 |
"step": 6000
|
| 895 |
},
|
| 896 |
{
|
| 897 |
"epoch": 32.0,
|
| 898 |
-
"eval_accuracy": 0.
|
| 899 |
-
"eval_loss":
|
| 900 |
-
"eval_runtime":
|
| 901 |
-
"eval_samples_per_second":
|
| 902 |
-
"eval_steps_per_second":
|
| 903 |
"step": 6000
|
| 904 |
},
|
| 905 |
{
|
| 906 |
"epoch": 32.0,
|
| 907 |
-
"eval_exact_match":
|
| 908 |
-
"eval_f1":
|
| 909 |
"step": 6000
|
| 910 |
},
|
| 911 |
{
|
| 912 |
-
"epoch": 32.
|
| 913 |
-
"grad_norm": 0.5082954168319702,
|
| 914 |
"learning_rate": 0.0003,
|
| 915 |
-
"loss": 0.
|
| 916 |
"step": 6100
|
| 917 |
},
|
| 918 |
{
|
| 919 |
-
"epoch":
|
| 920 |
-
"eval_accuracy": 0.
|
| 921 |
-
"eval_loss": 2.
|
| 922 |
-
"eval_runtime":
|
| 923 |
-
"eval_samples_per_second":
|
| 924 |
-
"eval_steps_per_second":
|
| 925 |
"step": 6187
|
| 926 |
},
|
| 927 |
{
|
| 928 |
-
"epoch":
|
| 929 |
-
"eval_exact_match":
|
| 930 |
-
"eval_f1":
|
| 931 |
"step": 6187
|
| 932 |
},
|
| 933 |
{
|
| 934 |
-
"epoch": 33.
|
| 935 |
-
"grad_norm": 0.5624931454658508,
|
| 936 |
"learning_rate": 0.0003,
|
| 937 |
-
"loss": 0.
|
| 938 |
"step": 6200
|
| 939 |
},
|
| 940 |
{
|
| 941 |
"epoch": 33.6,
|
| 942 |
-
"grad_norm": 0.44979730248451233,
|
| 943 |
"learning_rate": 0.0003,
|
| 944 |
-
"loss": 0.
|
| 945 |
"step": 6300
|
| 946 |
},
|
| 947 |
{
|
| 948 |
"epoch": 34.0,
|
| 949 |
-
"eval_accuracy": 0.
|
| 950 |
-
"eval_loss":
|
| 951 |
-
"eval_runtime":
|
| 952 |
-
"eval_samples_per_second":
|
| 953 |
-
"eval_steps_per_second":
|
| 954 |
"step": 6375
|
| 955 |
},
|
| 956 |
{
|
| 957 |
"epoch": 34.0,
|
| 958 |
-
"eval_exact_match":
|
| 959 |
-
"eval_f1":
|
| 960 |
"step": 6375
|
| 961 |
},
|
| 962 |
{
|
| 963 |
-
"epoch": 34.
|
| 964 |
-
"grad_norm": 0.34029239416122437,
|
| 965 |
"learning_rate": 0.0003,
|
| 966 |
-
"loss": 0.
|
| 967 |
"step": 6400
|
| 968 |
},
|
| 969 |
{
|
| 970 |
-
"epoch": 34.
|
| 971 |
-
"grad_norm": 0.3919186294078827,
|
| 972 |
"learning_rate": 0.0003,
|
| 973 |
-
"loss": 0.
|
| 974 |
"step": 6500
|
| 975 |
},
|
| 976 |
{
|
| 977 |
-
"epoch":
|
| 978 |
-
"eval_accuracy": 0.
|
| 979 |
-
"eval_loss":
|
| 980 |
-
"eval_runtime":
|
| 981 |
-
"eval_samples_per_second":
|
| 982 |
-
"eval_steps_per_second":
|
| 983 |
"step": 6562
|
| 984 |
},
|
| 985 |
{
|
| 986 |
-
"epoch":
|
| 987 |
-
"eval_exact_match":
|
| 988 |
-
"eval_f1":
|
| 989 |
"step": 6562
|
| 990 |
},
|
| 991 |
{
|
| 992 |
"epoch": 35.2,
|
| 993 |
-
"grad_norm": 0.31261807680130005,
|
| 994 |
"learning_rate": 0.0003,
|
| 995 |
-
"loss": 0.
|
| 996 |
"step": 6600
|
| 997 |
},
|
| 998 |
{
|
| 999 |
-
"epoch": 35.
|
| 1000 |
-
"grad_norm": 0.37774205207824707,
|
| 1001 |
"learning_rate": 0.0003,
|
| 1002 |
-
"loss":
|
| 1003 |
"step": 6700
|
| 1004 |
},
|
| 1005 |
{
|
| 1006 |
"epoch": 36.0,
|
| 1007 |
-
"eval_accuracy": 0.
|
| 1008 |
-
"eval_loss":
|
| 1009 |
-
"eval_runtime":
|
| 1010 |
-
"eval_samples_per_second":
|
| 1011 |
-
"eval_steps_per_second":
|
| 1012 |
"step": 6750
|
| 1013 |
},
|
| 1014 |
{
|
| 1015 |
"epoch": 36.0,
|
| 1016 |
-
"eval_exact_match":
|
| 1017 |
-
"eval_f1":
|
| 1018 |
"step": 6750
|
| 1019 |
}
|
| 1020 |
],
|
| 1021 |
"logging_steps": 100,
|
| 1022 |
"max_steps": 9350,
|
| 1023 |
-
"num_input_tokens_seen": 0,
|
| 1024 |
"num_train_epochs": 50,
|
| 1025 |
"save_steps": 500,
|
| 1026 |
-
"
|
| 1027 |
-
"TrainerControl": {
|
| 1028 |
-
"args": {
|
| 1029 |
-
"should_epoch_stop": false,
|
| 1030 |
-
"should_evaluate": false,
|
| 1031 |
-
"should_log": false,
|
| 1032 |
-
"should_save": true,
|
| 1033 |
-
"should_training_stop": false
|
| 1034 |
-
},
|
| 1035 |
-
"attributes": {}
|
| 1036 |
-
}
|
| 1037 |
-
},
|
| 1038 |
-
"total_flos": 1.1755663888954163e+17,
|
| 1039 |
-
"train_batch_size": 1,
|
| 1040 |
"trial_name": null,
|
| 1041 |
"trial_params": null
|
| 1042 |
}
|
|
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.53,
|
|
|
|
| 13 |
"learning_rate": 0.0003,
|
| 14 |
+
"loss": 1.4433,
|
| 15 |
"step": 100
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 1.0,
|
| 19 |
+
"eval_accuracy": 0.6165641025641025,
|
| 20 |
+
"eval_loss": 1.2625622749328613,
|
| 21 |
+
"eval_runtime": 4.3923,
|
| 22 |
+
"eval_samples_per_second": 113.835,
|
| 23 |
+
"eval_steps_per_second": 14.343,
|
| 24 |
"step": 187
|
| 25 |
},
|
| 26 |
{
|
| 27 |
+
"epoch": 1.0,
|
| 28 |
+
"eval_exact_match": 26.4,
|
| 29 |
+
"eval_f1": 39.52571428571429,
|
| 30 |
"step": 187
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 1.07,
|
|
|
|
| 34 |
"learning_rate": 0.0003,
|
| 35 |
+
"loss": 1.2659,
|
| 36 |
"step": 200
|
| 37 |
},
|
| 38 |
{
|
| 39 |
"epoch": 1.6,
|
|
|
|
| 40 |
"learning_rate": 0.0003,
|
| 41 |
+
"loss": 1.0761,
|
| 42 |
"step": 300
|
| 43 |
},
|
| 44 |
{
|
| 45 |
"epoch": 2.0,
|
| 46 |
+
"eval_accuracy": 0.615897435897436,
|
| 47 |
+
"eval_loss": 1.295995831489563,
|
| 48 |
+
"eval_runtime": 4.4145,
|
| 49 |
+
"eval_samples_per_second": 113.264,
|
| 50 |
+
"eval_steps_per_second": 14.271,
|
| 51 |
"step": 375
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 2.0,
|
| 55 |
+
"eval_exact_match": 25.4,
|
| 56 |
+
"eval_f1": 38.36603174603176,
|
| 57 |
"step": 375
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 2.13,
|
|
|
|
| 61 |
"learning_rate": 0.0003,
|
| 62 |
+
"loss": 0.9947,
|
| 63 |
"step": 400
|
| 64 |
},
|
| 65 |
{
|
| 66 |
+
"epoch": 2.67,
|
|
|
|
| 67 |
"learning_rate": 0.0003,
|
| 68 |
+
"loss": 0.7909,
|
| 69 |
"step": 500
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 3.0,
|
| 73 |
+
"eval_accuracy": 0.6093846153846154,
|
| 74 |
+
"eval_loss": 1.42075777053833,
|
| 75 |
+
"eval_runtime": 4.4832,
|
| 76 |
+
"eval_samples_per_second": 111.529,
|
| 77 |
+
"eval_steps_per_second": 14.053,
|
| 78 |
"step": 562
|
| 79 |
},
|
| 80 |
{
|
| 81 |
+
"epoch": 3.0,
|
| 82 |
+
"eval_exact_match": 23.6,
|
| 83 |
+
"eval_f1": 36.76761904761906,
|
| 84 |
"step": 562
|
| 85 |
},
|
| 86 |
{
|
| 87 |
"epoch": 3.2,
|
|
|
|
| 88 |
"learning_rate": 0.0003,
|
| 89 |
+
"loss": 0.7343,
|
| 90 |
"step": 600
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"epoch": 3.73,
|
|
|
|
| 94 |
"learning_rate": 0.0003,
|
| 95 |
+
"loss": 0.6117,
|
| 96 |
"step": 700
|
| 97 |
},
|
| 98 |
{
|
| 99 |
"epoch": 4.0,
|
| 100 |
+
"eval_accuracy": 0.6065128205128205,
|
| 101 |
+
"eval_loss": 1.509896159172058,
|
| 102 |
+
"eval_runtime": 4.8185,
|
| 103 |
+
"eval_samples_per_second": 103.766,
|
| 104 |
+
"eval_steps_per_second": 13.075,
|
| 105 |
"step": 750
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 4.0,
|
| 109 |
+
"eval_exact_match": 22.6,
|
| 110 |
+
"eval_f1": 35.788253968253976,
|
| 111 |
"step": 750
|
| 112 |
},
|
| 113 |
{
|
| 114 |
+
"epoch": 4.27,
|
|
|
|
| 115 |
"learning_rate": 0.0003,
|
| 116 |
+
"loss": 0.5688,
|
| 117 |
"step": 800
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 4.8,
|
|
|
|
| 121 |
"learning_rate": 0.0003,
|
| 122 |
+
"loss": 0.535,
|
| 123 |
"step": 900
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 5.0,
|
| 127 |
+
"eval_accuracy": 0.6073333333333333,
|
| 128 |
+
"eval_loss": 1.6383951902389526,
|
| 129 |
+
"eval_runtime": 4.1044,
|
| 130 |
+
"eval_samples_per_second": 121.82,
|
| 131 |
+
"eval_steps_per_second": 15.349,
|
| 132 |
"step": 937
|
| 133 |
},
|
| 134 |
{
|
| 135 |
+
"epoch": 5.0,
|
| 136 |
+
"eval_exact_match": 23.8,
|
| 137 |
+
"eval_f1": 37.04809523809523,
|
| 138 |
"step": 937
|
| 139 |
},
|
| 140 |
{
|
| 141 |
+
"epoch": 5.33,
|
|
|
|
| 142 |
"learning_rate": 0.0003,
|
| 143 |
+
"loss": 0.4997,
|
| 144 |
"step": 1000
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 5.87,
|
|
|
|
| 148 |
"learning_rate": 0.0003,
|
| 149 |
+
"loss": 0.4917,
|
| 150 |
"step": 1100
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 6.0,
|
| 154 |
+
"eval_accuracy": 0.6064102564102564,
|
| 155 |
+
"eval_loss": 1.6917529106140137,
|
| 156 |
+
"eval_runtime": 4.8299,
|
| 157 |
+
"eval_samples_per_second": 103.522,
|
| 158 |
+
"eval_steps_per_second": 13.044,
|
| 159 |
"step": 1125
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 6.0,
|
| 163 |
+
"eval_exact_match": 24.2,
|
| 164 |
+
"eval_f1": 36.030317460317455,
|
| 165 |
"step": 1125
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 6.4,
|
|
|
|
| 169 |
"learning_rate": 0.0003,
|
| 170 |
+
"loss": 0.4605,
|
| 171 |
"step": 1200
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 6.93,
|
|
|
|
| 175 |
"learning_rate": 0.0003,
|
| 176 |
+
"loss": 0.4746,
|
| 177 |
"step": 1300
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 7.0,
|
| 181 |
+
"eval_accuracy": 0.6048205128205129,
|
| 182 |
+
"eval_loss": 1.7547425031661987,
|
| 183 |
+
"eval_runtime": 4.2021,
|
| 184 |
+
"eval_samples_per_second": 118.988,
|
| 185 |
+
"eval_steps_per_second": 14.993,
|
| 186 |
"step": 1312
|
| 187 |
},
|
| 188 |
{
|
| 189 |
+
"epoch": 7.0,
|
| 190 |
+
"eval_exact_match": 23.8,
|
| 191 |
+
"eval_f1": 36.49507936507937,
|
| 192 |
"step": 1312
|
| 193 |
},
|
| 194 |
{
|
| 195 |
+
"epoch": 7.47,
|
|
|
|
| 196 |
"learning_rate": 0.0003,
|
| 197 |
+
"loss": 0.4432,
|
| 198 |
"step": 1400
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 8.0,
|
|
|
|
| 202 |
"learning_rate": 0.0003,
|
| 203 |
+
"loss": 0.463,
|
| 204 |
"step": 1500
|
| 205 |
},
|
| 206 |
{
|
| 207 |
"epoch": 8.0,
|
| 208 |
+
"eval_accuracy": 0.6056923076923076,
|
| 209 |
+
"eval_loss": 1.7707523107528687,
|
| 210 |
+
"eval_runtime": 4.821,
|
| 211 |
+
"eval_samples_per_second": 103.713,
|
| 212 |
+
"eval_steps_per_second": 13.068,
|
| 213 |
"step": 1500
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 8.0,
|
| 217 |
+
"eval_exact_match": 22.4,
|
| 218 |
+
"eval_f1": 35.834603174603174,
|
| 219 |
"step": 1500
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 8.53,
|
|
|
|
| 223 |
"learning_rate": 0.0003,
|
| 224 |
+
"loss": 0.4307,
|
| 225 |
"step": 1600
|
| 226 |
},
|
| 227 |
{
|
| 228 |
+
"epoch": 9.0,
|
| 229 |
+
"eval_accuracy": 0.6051282051282051,
|
| 230 |
+
"eval_loss": 1.7772654294967651,
|
| 231 |
+
"eval_runtime": 4.4303,
|
| 232 |
+
"eval_samples_per_second": 112.86,
|
| 233 |
+
"eval_steps_per_second": 14.22,
|
| 234 |
"step": 1687
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 9.0,
|
| 238 |
+
"eval_exact_match": 24.4,
|
| 239 |
+
"eval_f1": 36.40761904761904,
|
| 240 |
"step": 1687
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 9.07,
|
|
|
|
| 244 |
"learning_rate": 0.0003,
|
| 245 |
+
"loss": 0.4694,
|
| 246 |
"step": 1700
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 9.6,
|
|
|
|
| 250 |
"learning_rate": 0.0003,
|
| 251 |
+
"loss": 0.4323,
|
| 252 |
"step": 1800
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 10.0,
|
| 256 |
+
"eval_accuracy": 0.5996923076923077,
|
| 257 |
+
"eval_loss": 1.7927325963974,
|
| 258 |
+
"eval_runtime": 4.4251,
|
| 259 |
+
"eval_samples_per_second": 112.993,
|
| 260 |
+
"eval_steps_per_second": 14.237,
|
| 261 |
"step": 1875
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 10.0,
|
| 265 |
+
"eval_exact_match": 23.2,
|
| 266 |
+
"eval_f1": 35.76095238095239,
|
| 267 |
"step": 1875
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 10.13,
|
|
|
|
| 271 |
"learning_rate": 0.0003,
|
| 272 |
+
"loss": 0.4392,
|
| 273 |
"step": 1900
|
| 274 |
},
|
| 275 |
{
|
| 276 |
+
"epoch": 10.67,
|
|
|
|
| 277 |
"learning_rate": 0.0003,
|
| 278 |
+
"loss": 0.4301,
|
| 279 |
"step": 2000
|
| 280 |
},
|
| 281 |
{
|
| 282 |
+
"epoch": 11.0,
|
| 283 |
+
"eval_accuracy": 0.6046153846153847,
|
| 284 |
+
"eval_loss": 1.820131778717041,
|
| 285 |
+
"eval_runtime": 4.2028,
|
| 286 |
+
"eval_samples_per_second": 118.968,
|
| 287 |
+
"eval_steps_per_second": 14.99,
|
| 288 |
"step": 2062
|
| 289 |
},
|
| 290 |
{
|
| 291 |
+
"epoch": 11.0,
|
| 292 |
+
"eval_exact_match": 22.6,
|
| 293 |
+
"eval_f1": 36.275425685425695,
|
| 294 |
"step": 2062
|
| 295 |
},
|
| 296 |
{
|
| 297 |
"epoch": 11.2,
|
|
|
|
| 298 |
"learning_rate": 0.0003,
|
| 299 |
+
"loss": 0.4325,
|
| 300 |
"step": 2100
|
| 301 |
},
|
| 302 |
{
|
| 303 |
+
"epoch": 11.73,
|
|
|
|
| 304 |
"learning_rate": 0.0003,
|
| 305 |
+
"loss": 0.4331,
|
| 306 |
"step": 2200
|
| 307 |
},
|
| 308 |
{
|
| 309 |
"epoch": 12.0,
|
| 310 |
+
"eval_accuracy": 0.6050769230769231,
|
| 311 |
+
"eval_loss": 1.8902928829193115,
|
| 312 |
+
"eval_runtime": 4.1616,
|
| 313 |
+
"eval_samples_per_second": 120.145,
|
| 314 |
+
"eval_steps_per_second": 15.138,
|
| 315 |
"step": 2250
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"epoch": 12.0,
|
| 319 |
+
"eval_exact_match": 23.6,
|
| 320 |
+
"eval_f1": 35.66952380952382,
|
| 321 |
"step": 2250
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"epoch": 12.27,
|
|
|
|
| 325 |
"learning_rate": 0.0003,
|
| 326 |
+
"loss": 0.43,
|
| 327 |
"step": 2300
|
| 328 |
},
|
| 329 |
{
|
| 330 |
"epoch": 12.8,
|
|
|
|
| 331 |
"learning_rate": 0.0003,
|
| 332 |
+
"loss": 0.4338,
|
| 333 |
"step": 2400
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 13.0,
|
| 337 |
+
"eval_accuracy": 0.6035384615384616,
|
| 338 |
+
"eval_loss": 1.8489463329315186,
|
| 339 |
+
"eval_runtime": 4.4116,
|
| 340 |
+
"eval_samples_per_second": 113.338,
|
| 341 |
+
"eval_steps_per_second": 14.281,
|
| 342 |
"step": 2437
|
| 343 |
},
|
| 344 |
{
|
| 345 |
+
"epoch": 13.0,
|
| 346 |
+
"eval_exact_match": 19.6,
|
| 347 |
+
"eval_f1": 33.558888888888895,
|
| 348 |
"step": 2437
|
| 349 |
},
|
| 350 |
{
|
| 351 |
+
"epoch": 13.33,
|
|
|
|
| 352 |
"learning_rate": 0.0003,
|
| 353 |
+
"loss": 0.4261,
|
| 354 |
"step": 2500
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 13.87,
|
|
|
|
| 358 |
"learning_rate": 0.0003,
|
| 359 |
+
"loss": 0.4331,
|
| 360 |
"step": 2600
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 14.0,
|
| 364 |
+
"eval_accuracy": 0.6024102564102564,
|
| 365 |
+
"eval_loss": 1.9057968854904175,
|
| 366 |
+
"eval_runtime": 4.5305,
|
| 367 |
+
"eval_samples_per_second": 110.363,
|
| 368 |
+
"eval_steps_per_second": 13.906,
|
| 369 |
"step": 2625
|
| 370 |
},
|
| 371 |
{
|
| 372 |
"epoch": 14.0,
|
| 373 |
+
"eval_exact_match": 23.0,
|
| 374 |
+
"eval_f1": 34.747936507936515,
|
| 375 |
"step": 2625
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 14.4,
|
|
|
|
| 379 |
"learning_rate": 0.0003,
|
| 380 |
+
"loss": 0.4247,
|
| 381 |
"step": 2700
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 14.93,
|
|
|
|
| 385 |
"learning_rate": 0.0003,
|
| 386 |
+
"loss": 0.4362,
|
| 387 |
"step": 2800
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 15.0,
|
| 391 |
+
"eval_accuracy": 0.6004102564102564,
|
| 392 |
+
"eval_loss": 1.8960095643997192,
|
| 393 |
+
"eval_runtime": 4.4093,
|
| 394 |
+
"eval_samples_per_second": 113.396,
|
| 395 |
+
"eval_steps_per_second": 14.288,
|
| 396 |
"step": 2812
|
| 397 |
},
|
| 398 |
{
|
| 399 |
+
"epoch": 15.0,
|
| 400 |
+
"eval_exact_match": 12.2,
|
| 401 |
+
"eval_f1": 20.72142857142858,
|
| 402 |
"step": 2812
|
| 403 |
},
|
| 404 |
{
|
| 405 |
+
"epoch": 15.47,
|
|
|
|
| 406 |
"learning_rate": 0.0003,
|
| 407 |
+
"loss": 0.4188,
|
| 408 |
"step": 2900
|
| 409 |
},
|
| 410 |
{
|
| 411 |
"epoch": 16.0,
|
|
|
|
| 412 |
"learning_rate": 0.0003,
|
| 413 |
+
"loss": 0.4371,
|
| 414 |
"step": 3000
|
| 415 |
},
|
| 416 |
{
|
| 417 |
"epoch": 16.0,
|
| 418 |
+
"eval_accuracy": 0.6011794871794872,
|
| 419 |
+
"eval_loss": 1.9080209732055664,
|
| 420 |
+
"eval_runtime": 4.8301,
|
| 421 |
+
"eval_samples_per_second": 103.517,
|
| 422 |
+
"eval_steps_per_second": 13.043,
|
| 423 |
"step": 3000
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"epoch": 16.0,
|
| 427 |
+
"eval_exact_match": 21.6,
|
| 428 |
+
"eval_f1": 33.10307692307693,
|
| 429 |
"step": 3000
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 16.53,
|
|
|
|
| 433 |
"learning_rate": 0.0003,
|
| 434 |
+
"loss": 0.4153,
|
| 435 |
"step": 3100
|
| 436 |
},
|
| 437 |
{
|
| 438 |
+
"epoch": 17.0,
|
| 439 |
+
"eval_accuracy": 0.601948717948718,
|
| 440 |
+
"eval_loss": 1.9089771509170532,
|
| 441 |
+
"eval_runtime": 4.3169,
|
| 442 |
+
"eval_samples_per_second": 115.823,
|
| 443 |
+
"eval_steps_per_second": 14.594,
|
| 444 |
"step": 3187
|
| 445 |
},
|
| 446 |
{
|
| 447 |
+
"epoch": 17.0,
|
| 448 |
+
"eval_exact_match": 22.8,
|
| 449 |
+
"eval_f1": 34.82815628815629,
|
| 450 |
"step": 3187
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 17.07,
|
|
|
|
| 454 |
"learning_rate": 0.0003,
|
| 455 |
+
"loss": 0.4285,
|
| 456 |
"step": 3200
|
| 457 |
},
|
| 458 |
{
|
| 459 |
"epoch": 17.6,
|
|
|
|
| 460 |
"learning_rate": 0.0003,
|
| 461 |
+
"loss": 0.4151,
|
| 462 |
"step": 3300
|
| 463 |
},
|
| 464 |
{
|
| 465 |
"epoch": 18.0,
|
| 466 |
+
"eval_accuracy": 0.6003589743589743,
|
| 467 |
+
"eval_loss": 1.913118839263916,
|
| 468 |
+
"eval_runtime": 4.2766,
|
| 469 |
+
"eval_samples_per_second": 116.915,
|
| 470 |
+
"eval_steps_per_second": 14.731,
|
| 471 |
"step": 3375
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 18.0,
|
| 475 |
+
"eval_exact_match": 0.0,
|
| 476 |
+
"eval_f1": 0.15157160620821333,
|
| 477 |
"step": 3375
|
| 478 |
},
|
| 479 |
{
|
| 480 |
+
"epoch": 18.13,
|
|
|
|
| 481 |
"learning_rate": 0.0003,
|
| 482 |
+
"loss": 0.4262,
|
| 483 |
"step": 3400
|
| 484 |
},
|
| 485 |
{
|
| 486 |
+
"epoch": 18.67,
|
|
|
|
| 487 |
"learning_rate": 0.0003,
|
| 488 |
+
"loss": 0.426,
|
| 489 |
"step": 3500
|
| 490 |
},
|
| 491 |
{
|
| 492 |
+
"epoch": 19.0,
|
| 493 |
+
"eval_accuracy": 0.574974358974359,
|
| 494 |
+
"eval_loss": 2.0016062259674072,
|
| 495 |
+
"eval_runtime": 4.121,
|
| 496 |
+
"eval_samples_per_second": 121.33,
|
| 497 |
+
"eval_steps_per_second": 15.288,
|
| 498 |
"step": 3562
|
| 499 |
},
|
| 500 |
{
|
| 501 |
+
"epoch": 19.0,
|
| 502 |
+
"eval_exact_match": 0.0,
|
| 503 |
+
"eval_f1": 0.0,
|
| 504 |
"step": 3562
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 19.2,
|
|
|
|
| 508 |
"learning_rate": 0.0003,
|
| 509 |
+
"loss": 2.0542,
|
| 510 |
"step": 3600
|
| 511 |
},
|
| 512 |
{
|
| 513 |
+
"epoch": 19.73,
|
|
|
|
| 514 |
"learning_rate": 0.0003,
|
| 515 |
+
"loss": 0.4475,
|
| 516 |
"step": 3700
|
| 517 |
},
|
| 518 |
{
|
| 519 |
"epoch": 20.0,
|
| 520 |
+
"eval_accuracy": 0.600974358974359,
|
| 521 |
+
"eval_loss": 1.887317419052124,
|
| 522 |
+
"eval_runtime": 4.319,
|
| 523 |
+
"eval_samples_per_second": 115.767,
|
| 524 |
+
"eval_steps_per_second": 14.587,
|
| 525 |
"step": 3750
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 20.0,
|
| 529 |
+
"eval_exact_match": 6.8,
|
| 530 |
+
"eval_f1": 11.268571428571423,
|
| 531 |
"step": 3750
|
| 532 |
},
|
| 533 |
{
|
| 534 |
+
"epoch": 20.27,
|
|
|
|
| 535 |
"learning_rate": 0.0003,
|
| 536 |
+
"loss": 0.4227,
|
| 537 |
"step": 3800
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 20.8,
|
|
|
|
| 541 |
"learning_rate": 0.0003,
|
| 542 |
+
"loss": 0.4325,
|
| 543 |
"step": 3900
|
| 544 |
},
|
| 545 |
{
|
| 546 |
+
"epoch": 21.0,
|
| 547 |
+
"eval_accuracy": 0.6008717948717949,
|
| 548 |
+
"eval_loss": 1.835098147392273,
|
| 549 |
+
"eval_runtime": 4.2241,
|
| 550 |
+
"eval_samples_per_second": 118.368,
|
| 551 |
+
"eval_steps_per_second": 14.914,
|
| 552 |
"step": 3937
|
| 553 |
},
|
| 554 |
{
|
| 555 |
+
"epoch": 21.0,
|
| 556 |
+
"eval_exact_match": 0.0,
|
| 557 |
+
"eval_f1": 0.295334700387332,
|
| 558 |
"step": 3937
|
| 559 |
},
|
| 560 |
{
|
| 561 |
+
"epoch": 21.33,
|
|
|
|
| 562 |
"learning_rate": 0.0003,
|
| 563 |
+
"loss": 0.4318,
|
| 564 |
"step": 4000
|
| 565 |
},
|
| 566 |
{
|
| 567 |
+
"epoch": 21.87,
|
|
|
|
| 568 |
"learning_rate": 0.0003,
|
| 569 |
+
"loss": 0.4542,
|
| 570 |
"step": 4100
|
| 571 |
},
|
| 572 |
{
|
| 573 |
"epoch": 22.0,
|
| 574 |
+
"eval_accuracy": 0.598974358974359,
|
| 575 |
+
"eval_loss": 1.865936040878296,
|
| 576 |
+
"eval_runtime": 4.4112,
|
| 577 |
+
"eval_samples_per_second": 113.347,
|
| 578 |
+
"eval_steps_per_second": 14.282,
|
| 579 |
"step": 4125
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 22.0,
|
| 583 |
+
"eval_exact_match": 6.6,
|
| 584 |
+
"eval_f1": 10.301587301587299,
|
| 585 |
"step": 4125
|
| 586 |
},
|
| 587 |
{
|
| 588 |
"epoch": 22.4,
|
|
|
|
| 589 |
"learning_rate": 0.0003,
|
| 590 |
+
"loss": 1.4938,
|
| 591 |
"step": 4200
|
| 592 |
},
|
| 593 |
{
|
| 594 |
+
"epoch": 22.93,
|
|
|
|
| 595 |
"learning_rate": 0.0003,
|
| 596 |
+
"loss": 0.5662,
|
| 597 |
"step": 4300
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 23.0,
|
| 601 |
+
"eval_accuracy": 0.5963589743589743,
|
| 602 |
+
"eval_loss": 1.8427150249481201,
|
| 603 |
+
"eval_runtime": 4.8431,
|
| 604 |
+
"eval_samples_per_second": 103.239,
|
| 605 |
+
"eval_steps_per_second": 13.008,
|
| 606 |
"step": 4312
|
| 607 |
},
|
| 608 |
{
|
| 609 |
+
"epoch": 23.0,
|
| 610 |
+
"eval_exact_match": 0.0,
|
| 611 |
+
"eval_f1": 0.13841036414565827,
|
| 612 |
"step": 4312
|
| 613 |
},
|
| 614 |
{
|
| 615 |
+
"epoch": 23.47,
|
|
|
|
| 616 |
"learning_rate": 0.0003,
|
| 617 |
+
"loss": 0.4186,
|
| 618 |
"step": 4400
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 24.0,
|
|
|
|
| 622 |
"learning_rate": 0.0003,
|
| 623 |
+
"loss": 0.4365,
|
| 624 |
"step": 4500
|
| 625 |
},
|
| 626 |
{
|
| 627 |
"epoch": 24.0,
|
| 628 |
+
"eval_accuracy": 0.5993333333333334,
|
| 629 |
+
"eval_loss": 1.8092677593231201,
|
| 630 |
+
"eval_runtime": 4.206,
|
| 631 |
+
"eval_samples_per_second": 118.879,
|
| 632 |
+
"eval_steps_per_second": 14.979,
|
| 633 |
"step": 4500
|
| 634 |
},
|
| 635 |
{
|
| 636 |
"epoch": 24.0,
|
| 637 |
+
"eval_exact_match": 1.2,
|
| 638 |
+
"eval_f1": 4.495396065922379,
|
| 639 |
"step": 4500
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 24.53,
|
|
|
|
| 643 |
"learning_rate": 0.0003,
|
| 644 |
+
"loss": 0.4153,
|
| 645 |
"step": 4600
|
| 646 |
},
|
| 647 |
{
|
| 648 |
+
"epoch": 25.0,
|
| 649 |
+
"eval_accuracy": 0.6022051282051282,
|
| 650 |
+
"eval_loss": 1.841846227645874,
|
| 651 |
+
"eval_runtime": 4.618,
|
| 652 |
+
"eval_samples_per_second": 108.271,
|
| 653 |
+
"eval_steps_per_second": 13.642,
|
| 654 |
"step": 4687
|
| 655 |
},
|
| 656 |
{
|
| 657 |
+
"epoch": 25.0,
|
| 658 |
+
"eval_exact_match": 1.6,
|
| 659 |
+
"eval_f1": 2.857142857142857,
|
| 660 |
"step": 4687
|
| 661 |
},
|
| 662 |
{
|
| 663 |
+
"epoch": 25.07,
|
|
|
|
| 664 |
"learning_rate": 0.0003,
|
| 665 |
+
"loss": 0.4275,
|
| 666 |
"step": 4700
|
| 667 |
},
|
| 668 |
{
|
| 669 |
"epoch": 25.6,
|
|
|
|
| 670 |
"learning_rate": 0.0003,
|
| 671 |
+
"loss": 0.4099,
|
| 672 |
"step": 4800
|
| 673 |
},
|
| 674 |
{
|
| 675 |
"epoch": 26.0,
|
| 676 |
+
"eval_accuracy": 0.6001538461538461,
|
| 677 |
+
"eval_loss": 1.974199652671814,
|
| 678 |
+
"eval_runtime": 4.4307,
|
| 679 |
+
"eval_samples_per_second": 112.848,
|
| 680 |
+
"eval_steps_per_second": 14.219,
|
| 681 |
"step": 4875
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 26.0,
|
| 685 |
+
"eval_exact_match": 1.8,
|
| 686 |
+
"eval_f1": 2.9238095238095236,
|
| 687 |
"step": 4875
|
| 688 |
},
|
| 689 |
{
|
| 690 |
+
"epoch": 26.13,
|
|
|
|
| 691 |
"learning_rate": 0.0003,
|
| 692 |
+
"loss": 0.4165,
|
| 693 |
"step": 4900
|
| 694 |
},
|
| 695 |
{
|
| 696 |
+
"epoch": 26.67,
|
|
|
|
| 697 |
"learning_rate": 0.0003,
|
| 698 |
+
"loss": 0.4101,
|
| 699 |
"step": 5000
|
| 700 |
},
|
| 701 |
{
|
| 702 |
+
"epoch": 27.0,
|
| 703 |
+
"eval_accuracy": 0.5987179487179487,
|
| 704 |
+
"eval_loss": 1.968170166015625,
|
| 705 |
+
"eval_runtime": 4.4099,
|
| 706 |
+
"eval_samples_per_second": 113.38,
|
| 707 |
+
"eval_steps_per_second": 14.286,
|
| 708 |
"step": 5062
|
| 709 |
},
|
| 710 |
{
|
| 711 |
+
"epoch": 27.0,
|
| 712 |
+
"eval_exact_match": 1.4,
|
| 713 |
+
"eval_f1": 2.3733333333333335,
|
| 714 |
"step": 5062
|
| 715 |
},
|
| 716 |
{
|
| 717 |
"epoch": 27.2,
|
|
|
|
| 718 |
"learning_rate": 0.0003,
|
| 719 |
+
"loss": 0.4135,
|
| 720 |
"step": 5100
|
| 721 |
},
|
| 722 |
{
|
| 723 |
+
"epoch": 27.73,
|
|
|
|
| 724 |
"learning_rate": 0.0003,
|
| 725 |
+
"loss": 0.4122,
|
| 726 |
"step": 5200
|
| 727 |
},
|
| 728 |
{
|
| 729 |
"epoch": 28.0,
|
| 730 |
+
"eval_accuracy": 0.5991282051282051,
|
| 731 |
+
"eval_loss": 1.9906134605407715,
|
| 732 |
+
"eval_runtime": 4.1058,
|
| 733 |
+
"eval_samples_per_second": 121.778,
|
| 734 |
+
"eval_steps_per_second": 15.344,
|
| 735 |
"step": 5250
|
| 736 |
},
|
| 737 |
{
|
| 738 |
"epoch": 28.0,
|
| 739 |
+
"eval_exact_match": 0.0,
|
| 740 |
+
"eval_f1": 0.22539784029553853,
|
| 741 |
"step": 5250
|
| 742 |
},
|
| 743 |
{
|
| 744 |
+
"epoch": 28.27,
|
|
|
|
| 745 |
"learning_rate": 0.0003,
|
| 746 |
+
"loss": 0.4086,
|
| 747 |
"step": 5300
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 28.8,
|
|
|
|
| 751 |
"learning_rate": 0.0003,
|
| 752 |
+
"loss": 0.4116,
|
| 753 |
"step": 5400
|
| 754 |
},
|
| 755 |
{
|
| 756 |
+
"epoch": 29.0,
|
| 757 |
+
"eval_accuracy": 0.598974358974359,
|
| 758 |
+
"eval_loss": 1.9427579641342163,
|
| 759 |
+
"eval_runtime": 5.1461,
|
| 760 |
+
"eval_samples_per_second": 97.161,
|
| 761 |
+
"eval_steps_per_second": 12.242,
|
| 762 |
"step": 5437
|
| 763 |
},
|
| 764 |
{
|
| 765 |
+
"epoch": 29.0,
|
| 766 |
+
"eval_exact_match": 1.0,
|
| 767 |
+
"eval_f1": 1.2216666666666667,
|
| 768 |
"step": 5437
|
| 769 |
},
|
| 770 |
{
|
| 771 |
+
"epoch": 29.33,
|
|
|
|
| 772 |
"learning_rate": 0.0003,
|
| 773 |
+
"loss": 0.4066,
|
| 774 |
"step": 5500
|
| 775 |
},
|
| 776 |
{
|
| 777 |
+
"epoch": 29.87,
|
|
|
|
| 778 |
"learning_rate": 0.0003,
|
| 779 |
+
"loss": 0.4158,
|
| 780 |
"step": 5600
|
| 781 |
},
|
| 782 |
{
|
| 783 |
"epoch": 30.0,
|
| 784 |
+
"eval_accuracy": 0.5987179487179487,
|
| 785 |
+
"eval_loss": 1.9261677265167236,
|
| 786 |
+
"eval_runtime": 4.2556,
|
| 787 |
+
"eval_samples_per_second": 117.493,
|
| 788 |
+
"eval_steps_per_second": 14.804,
|
| 789 |
"step": 5625
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 30.0,
|
| 793 |
+
"eval_exact_match": 0.0,
|
| 794 |
+
"eval_f1": 0.0,
|
| 795 |
"step": 5625
|
| 796 |
},
|
| 797 |
{
|
| 798 |
"epoch": 30.4,
|
|
|
|
| 799 |
"learning_rate": 0.0003,
|
| 800 |
+
"loss": 0.4076,
|
| 801 |
"step": 5700
|
| 802 |
},
|
| 803 |
{
|
| 804 |
+
"epoch": 30.93,
|
|
|
|
| 805 |
"learning_rate": 0.0003,
|
| 806 |
+
"loss": 0.4215,
|
| 807 |
"step": 5800
|
| 808 |
},
|
| 809 |
{
|
| 810 |
+
"epoch": 31.0,
|
| 811 |
+
"eval_accuracy": 0.5961025641025641,
|
| 812 |
+
"eval_loss": 1.9734665155410767,
|
| 813 |
+
"eval_runtime": 4.4089,
|
| 814 |
+
"eval_samples_per_second": 113.407,
|
| 815 |
+
"eval_steps_per_second": 14.289,
|
| 816 |
"step": 5812
|
| 817 |
},
|
| 818 |
{
|
| 819 |
+
"epoch": 31.0,
|
| 820 |
+
"eval_exact_match": 12.2,
|
| 821 |
+
"eval_f1": 15.156392496392497,
|
| 822 |
"step": 5812
|
| 823 |
},
|
| 824 |
{
|
| 825 |
+
"epoch": 31.47,
|
|
|
|
| 826 |
"learning_rate": 0.0003,
|
| 827 |
+
"loss": 0.4092,
|
| 828 |
"step": 5900
|
| 829 |
},
|
| 830 |
{
|
| 831 |
"epoch": 32.0,
|
|
|
|
| 832 |
"learning_rate": 0.0003,
|
| 833 |
+
"loss": 0.4249,
|
| 834 |
"step": 6000
|
| 835 |
},
|
| 836 |
{
|
| 837 |
"epoch": 32.0,
|
| 838 |
+
"eval_accuracy": 0.5966153846153847,
|
| 839 |
+
"eval_loss": 1.9393236637115479,
|
| 840 |
+
"eval_runtime": 4.2041,
|
| 841 |
+
"eval_samples_per_second": 118.931,
|
| 842 |
+
"eval_steps_per_second": 14.985,
|
| 843 |
"step": 6000
|
| 844 |
},
|
| 845 |
{
|
| 846 |
"epoch": 32.0,
|
| 847 |
+
"eval_exact_match": 5.6,
|
| 848 |
+
"eval_f1": 7.9144444444444435,
|
| 849 |
"step": 6000
|
| 850 |
},
|
| 851 |
{
|
| 852 |
+
"epoch": 32.53,
|
|
|
|
| 853 |
"learning_rate": 0.0003,
|
| 854 |
+
"loss": 0.4049,
|
| 855 |
"step": 6100
|
| 856 |
},
|
| 857 |
{
|
| 858 |
+
"epoch": 33.0,
|
| 859 |
+
"eval_accuracy": 0.5955897435897436,
|
| 860 |
+
"eval_loss": 2.008298873901367,
|
| 861 |
+
"eval_runtime": 4.4306,
|
| 862 |
+
"eval_samples_per_second": 112.851,
|
| 863 |
+
"eval_steps_per_second": 14.219,
|
| 864 |
"step": 6187
|
| 865 |
},
|
| 866 |
{
|
| 867 |
+
"epoch": 33.0,
|
| 868 |
+
"eval_exact_match": 0.0,
|
| 869 |
+
"eval_f1": 0.0,
|
| 870 |
"step": 6187
|
| 871 |
},
|
| 872 |
{
|
| 873 |
+
"epoch": 33.07,
|
|
|
|
| 874 |
"learning_rate": 0.0003,
|
| 875 |
+
"loss": 0.4219,
|
| 876 |
"step": 6200
|
| 877 |
},
|
| 878 |
{
|
| 879 |
"epoch": 33.6,
|
|
|
|
| 880 |
"learning_rate": 0.0003,
|
| 881 |
+
"loss": 0.4077,
|
| 882 |
"step": 6300
|
| 883 |
},
|
| 884 |
{
|
| 885 |
"epoch": 34.0,
|
| 886 |
+
"eval_accuracy": 0.5962051282051282,
|
| 887 |
+
"eval_loss": 1.9472050666809082,
|
| 888 |
+
"eval_runtime": 4.1025,
|
| 889 |
+
"eval_samples_per_second": 121.878,
|
| 890 |
+
"eval_steps_per_second": 15.357,
|
| 891 |
"step": 6375
|
| 892 |
},
|
| 893 |
{
|
| 894 |
"epoch": 34.0,
|
| 895 |
+
"eval_exact_match": 1.0,
|
| 896 |
+
"eval_f1": 1.2115698923090228,
|
| 897 |
"step": 6375
|
| 898 |
},
|
| 899 |
{
|
| 900 |
+
"epoch": 34.13,
|
|
|
|
| 901 |
"learning_rate": 0.0003,
|
| 902 |
+
"loss": 0.4149,
|
| 903 |
"step": 6400
|
| 904 |
},
|
| 905 |
{
|
| 906 |
+
"epoch": 34.67,
|
|
|
|
| 907 |
"learning_rate": 0.0003,
|
| 908 |
+
"loss": 0.4078,
|
| 909 |
"step": 6500
|
| 910 |
},
|
| 911 |
{
|
| 912 |
+
"epoch": 35.0,
|
| 913 |
+
"eval_accuracy": 0.5944615384615385,
|
| 914 |
+
"eval_loss": 1.9796123504638672,
|
| 915 |
+
"eval_runtime": 4.4312,
|
| 916 |
+
"eval_samples_per_second": 112.836,
|
| 917 |
+
"eval_steps_per_second": 14.217,
|
| 918 |
"step": 6562
|
| 919 |
},
|
| 920 |
{
|
| 921 |
+
"epoch": 35.0,
|
| 922 |
+
"eval_exact_match": 12.4,
|
| 923 |
+
"eval_f1": 17.85968253968255,
|
| 924 |
"step": 6562
|
| 925 |
},
|
| 926 |
{
|
| 927 |
"epoch": 35.2,
|
|
|
|
| 928 |
"learning_rate": 0.0003,
|
| 929 |
+
"loss": 0.4115,
|
| 930 |
"step": 6600
|
| 931 |
},
|
| 932 |
{
|
| 933 |
+
"epoch": 35.73,
|
|
|
|
| 934 |
"learning_rate": 0.0003,
|
| 935 |
+
"loss": 2.7734,
|
| 936 |
"step": 6700
|
| 937 |
},
|
| 938 |
{
|
| 939 |
"epoch": 36.0,
|
| 940 |
+
"eval_accuracy": 0.3344615384615385,
|
| 941 |
+
"eval_loss": 5.233836650848389,
|
| 942 |
+
"eval_runtime": 4.8797,
|
| 943 |
+
"eval_samples_per_second": 102.465,
|
| 944 |
+
"eval_steps_per_second": 12.911,
|
| 945 |
"step": 6750
|
| 946 |
},
|
| 947 |
{
|
| 948 |
"epoch": 36.0,
|
| 949 |
+
"eval_exact_match": 0.0,
|
| 950 |
+
"eval_f1": 0.11876160079321695,
|
| 951 |
"step": 6750
|
| 952 |
}
|
| 953 |
],
|
| 954 |
"logging_steps": 100,
|
| 955 |
"max_steps": 9350,
|
|
|
|
| 956 |
"num_train_epochs": 50,
|
| 957 |
"save_steps": 500,
|
| 958 |
+
"total_flos": 2.518627057097769e+17,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 959 |
"trial_name": null,
|
| 960 |
"trial_params": null
|
| 961 |
}
|
checkpoint-6750/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:79a83a5d44cad230e338c4f4f6393fbd052223dad36807daa3baa278ceb8a412
|
| 3 |
+
size 4728
|