Training in progress, epoch 21, checkpoint
Browse files- checkpoint-5250/adapter_config.json +1 -1
- checkpoint-5250/adapter_model.bin +3 -0
- checkpoint-5250/added_tokens.json +3 -3
- checkpoint-5250/optimizer.pt +2 -2
- checkpoint-5250/rng_state_0.pth +1 -1
- checkpoint-5250/rng_state_1.pth +1 -1
- checkpoint-5250/rng_state_2.pth +1 -1
- checkpoint-5250/rng_state_3.pth +1 -1
- checkpoint-5250/special_tokens_map.json +4 -12
- checkpoint-5250/tokenizer.json +0 -0
- checkpoint-5250/tokenizer.model +3 -0
- checkpoint-5250/tokenizer_config.json +17 -20
- checkpoint-5250/trainer_state.json +200 -266
- checkpoint-5250/training_args.bin +2 -2
checkpoint-5250/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-5250/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7dd2f567d9f040aa6f1631daac5120c2e38bed19ba88aaa9eb187474804c4649
|
| 3 |
+
size 143269386
|
checkpoint-5250/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-5250/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 sha256:022e6b98e812cc70beea7c3f5ae6c2711ef450a9e30c7009478c25a1604c7dc6
|
| 3 |
+
size 286585234
|
checkpoint-5250/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:fa389a5c39813c33c94244e6c59998d5757826ad938c860b416b7b8e2bd143c8
|
| 3 |
size 15024
|
checkpoint-5250/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 sha256:5e101e7c4db99629f66e59ae640fa738187c8263d2f2a65d10e1b23c92ceb188
|
| 3 |
size 15024
|
checkpoint-5250/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 sha256:538a18d56e9b562ae5447167ef9342ae0f40a0c85c5bd353c51101d0d50d9e9b
|
| 3 |
size 15024
|
checkpoint-5250/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 sha256:409441e402052d7d2ddcc43ea552c7bd82ee5f22f16ae53bc360c7f920e814c3
|
| 3 |
size 15024
|
checkpoint-5250/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-5250/tokenizer.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-5250/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
checkpoint-5250/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-5250/trainer_state.json
CHANGED
|
@@ -10,703 +10,637 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.4,
|
| 13 |
-
"grad_norm": 0.5479037761688232,
|
| 14 |
"learning_rate": 0.0005,
|
| 15 |
-
"loss": 2.
|
| 16 |
"step": 100
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.8,
|
| 20 |
-
"grad_norm": 0.5442906618118286,
|
| 21 |
"learning_rate": 0.0005,
|
| 22 |
-
"loss":
|
| 23 |
"step": 200
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 1.0,
|
| 27 |
-
"eval_accuracy": 0.
|
| 28 |
-
"eval_loss": 2.
|
| 29 |
-
"eval_runtime":
|
| 30 |
-
"eval_samples_per_second":
|
| 31 |
-
"eval_steps_per_second":
|
| 32 |
"step": 250
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 1.0,
|
| 36 |
-
"eval_exact_match":
|
| 37 |
-
"eval_f1":
|
| 38 |
"step": 250
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.2,
|
| 42 |
-
"grad_norm": 0.6918729543685913,
|
| 43 |
"learning_rate": 0.0005,
|
| 44 |
-
"loss":
|
| 45 |
"step": 300
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 1.6,
|
| 49 |
-
"grad_norm": 0.7950774431228638,
|
| 50 |
"learning_rate": 0.0005,
|
| 51 |
-
"loss": 1.
|
| 52 |
"step": 400
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 2.0,
|
| 56 |
-
"grad_norm": 0.6913579106330872,
|
| 57 |
"learning_rate": 0.0005,
|
| 58 |
-
"loss": 1.
|
| 59 |
"step": 500
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 2.0,
|
| 63 |
-
"eval_accuracy": 0.
|
| 64 |
-
"eval_loss":
|
| 65 |
-
"eval_runtime":
|
| 66 |
-
"eval_samples_per_second":
|
| 67 |
-
"eval_steps_per_second":
|
| 68 |
"step": 500
|
| 69 |
},
|
| 70 |
{
|
| 71 |
"epoch": 2.0,
|
| 72 |
-
"eval_exact_match":
|
| 73 |
-
"eval_f1":
|
| 74 |
"step": 500
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 2.4,
|
| 78 |
-
"grad_norm": 0.9923084378242493,
|
| 79 |
"learning_rate": 0.0005,
|
| 80 |
-
"loss": 1.
|
| 81 |
"step": 600
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 2.8,
|
| 85 |
-
"grad_norm": 1.0025588274002075,
|
| 86 |
"learning_rate": 0.0005,
|
| 87 |
-
"loss": 1.
|
| 88 |
"step": 700
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 3.0,
|
| 92 |
-
"eval_accuracy": 0.
|
| 93 |
-
"eval_loss":
|
| 94 |
-
"eval_runtime":
|
| 95 |
-
"eval_samples_per_second":
|
| 96 |
-
"eval_steps_per_second":
|
| 97 |
"step": 750
|
| 98 |
},
|
| 99 |
{
|
| 100 |
"epoch": 3.0,
|
| 101 |
-
"eval_exact_match":
|
| 102 |
-
"eval_f1":
|
| 103 |
"step": 750
|
| 104 |
},
|
| 105 |
{
|
| 106 |
"epoch": 3.2,
|
| 107 |
-
"grad_norm": 1.0352345705032349,
|
| 108 |
"learning_rate": 0.0005,
|
| 109 |
-
"loss": 1.
|
| 110 |
"step": 800
|
| 111 |
},
|
| 112 |
{
|
| 113 |
"epoch": 3.6,
|
| 114 |
-
"grad_norm": 1.1078394651412964,
|
| 115 |
"learning_rate": 0.0005,
|
| 116 |
-
"loss": 0.
|
| 117 |
"step": 900
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 4.0,
|
| 121 |
-
"grad_norm": 1.14862060546875,
|
| 122 |
"learning_rate": 0.0005,
|
| 123 |
-
"loss":
|
| 124 |
"step": 1000
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 4.0,
|
| 128 |
-
"eval_accuracy": 0.
|
| 129 |
-
"eval_loss": 2.
|
| 130 |
-
"eval_runtime": 5.
|
| 131 |
-
"eval_samples_per_second":
|
| 132 |
-
"eval_steps_per_second": 12.
|
| 133 |
"step": 1000
|
| 134 |
},
|
| 135 |
{
|
| 136 |
"epoch": 4.0,
|
| 137 |
-
"eval_exact_match":
|
| 138 |
-
"eval_f1":
|
| 139 |
"step": 1000
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"epoch": 4.4,
|
| 143 |
-
"grad_norm": 1.2231959104537964,
|
| 144 |
"learning_rate": 0.0005,
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 1100
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 4.8,
|
| 150 |
-
"grad_norm": 1.1669212579727173,
|
| 151 |
"learning_rate": 0.0005,
|
| 152 |
-
"loss": 0.
|
| 153 |
"step": 1200
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 5.0,
|
| 157 |
-
"eval_accuracy": 0.
|
| 158 |
-
"eval_loss":
|
| 159 |
-
"eval_runtime":
|
| 160 |
-
"eval_samples_per_second":
|
| 161 |
-
"eval_steps_per_second":
|
| 162 |
"step": 1250
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 5.0,
|
| 166 |
-
"eval_exact_match":
|
| 167 |
-
"eval_f1":
|
| 168 |
"step": 1250
|
| 169 |
},
|
| 170 |
{
|
| 171 |
"epoch": 5.2,
|
| 172 |
-
"grad_norm": 0.9650686979293823,
|
| 173 |
"learning_rate": 0.0005,
|
| 174 |
-
"loss": 0.
|
| 175 |
"step": 1300
|
| 176 |
},
|
| 177 |
{
|
| 178 |
"epoch": 5.6,
|
| 179 |
-
"grad_norm": 0.9634979963302612,
|
| 180 |
"learning_rate": 0.0005,
|
| 181 |
-
"loss": 0.
|
| 182 |
"step": 1400
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 6.0,
|
| 186 |
-
"grad_norm": 1.0452007055282593,
|
| 187 |
"learning_rate": 0.0005,
|
| 188 |
-
"loss": 0.
|
| 189 |
"step": 1500
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 6.0,
|
| 193 |
-
"eval_accuracy": 0.
|
| 194 |
-
"eval_loss":
|
| 195 |
-
"eval_runtime": 5.
|
| 196 |
-
"eval_samples_per_second": 97.
|
| 197 |
-
"eval_steps_per_second": 12.
|
| 198 |
"step": 1500
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 6.0,
|
| 202 |
-
"eval_exact_match":
|
| 203 |
-
"eval_f1":
|
| 204 |
"step": 1500
|
| 205 |
},
|
| 206 |
{
|
| 207 |
"epoch": 6.4,
|
| 208 |
-
"grad_norm": 1.0834842920303345,
|
| 209 |
"learning_rate": 0.0005,
|
| 210 |
-
"loss": 0.
|
| 211 |
"step": 1600
|
| 212 |
},
|
| 213 |
{
|
| 214 |
"epoch": 6.8,
|
| 215 |
-
"grad_norm": 0.9649925231933594,
|
| 216 |
"learning_rate": 0.0005,
|
| 217 |
-
"loss": 0.
|
| 218 |
"step": 1700
|
| 219 |
},
|
| 220 |
{
|
| 221 |
"epoch": 7.0,
|
| 222 |
-
"eval_accuracy": 0.
|
| 223 |
-
"eval_loss":
|
| 224 |
-
"eval_runtime":
|
| 225 |
-
"eval_samples_per_second":
|
| 226 |
-
"eval_steps_per_second":
|
| 227 |
"step": 1750
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 7.0,
|
| 231 |
-
"eval_exact_match":
|
| 232 |
-
"eval_f1":
|
| 233 |
"step": 1750
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 7.2,
|
| 237 |
-
"grad_norm": 0.8583429455757141,
|
| 238 |
"learning_rate": 0.0005,
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 1800
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 7.6,
|
| 244 |
-
"grad_norm": 0.9595094919204712,
|
| 245 |
"learning_rate": 0.0005,
|
| 246 |
-
"loss": 0.
|
| 247 |
"step": 1900
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"epoch": 8.0,
|
| 251 |
-
"grad_norm": 0.9260773062705994,
|
| 252 |
"learning_rate": 0.0005,
|
| 253 |
-
"loss": 0.
|
| 254 |
"step": 2000
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 8.0,
|
| 258 |
-
"eval_accuracy": 0.
|
| 259 |
-
"eval_loss":
|
| 260 |
-
"eval_runtime":
|
| 261 |
-
"eval_samples_per_second":
|
| 262 |
-
"eval_steps_per_second":
|
| 263 |
"step": 2000
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 8.0,
|
| 267 |
-
"eval_exact_match":
|
| 268 |
-
"eval_f1":
|
| 269 |
"step": 2000
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 8.4,
|
| 273 |
-
"grad_norm": 1.045862078666687,
|
| 274 |
"learning_rate": 0.0005,
|
| 275 |
-
"loss": 0.
|
| 276 |
"step": 2100
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 8.8,
|
| 280 |
-
"grad_norm": 0.7850492000579834,
|
| 281 |
"learning_rate": 0.0005,
|
| 282 |
-
"loss": 0.
|
| 283 |
"step": 2200
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 9.0,
|
| 287 |
-
"eval_accuracy": 0.
|
| 288 |
-
"eval_loss":
|
| 289 |
-
"eval_runtime":
|
| 290 |
-
"eval_samples_per_second":
|
| 291 |
-
"eval_steps_per_second":
|
| 292 |
"step": 2250
|
| 293 |
},
|
| 294 |
{
|
| 295 |
"epoch": 9.0,
|
| 296 |
-
"eval_exact_match":
|
| 297 |
-
"eval_f1":
|
| 298 |
"step": 2250
|
| 299 |
},
|
| 300 |
{
|
| 301 |
"epoch": 9.2,
|
| 302 |
-
"grad_norm": 0.8059192895889282,
|
| 303 |
"learning_rate": 0.0005,
|
| 304 |
-
"loss": 0.
|
| 305 |
"step": 2300
|
| 306 |
},
|
| 307 |
{
|
| 308 |
"epoch": 9.6,
|
| 309 |
-
"grad_norm": 0.9573262333869934,
|
| 310 |
"learning_rate": 0.0005,
|
| 311 |
-
"loss": 0.
|
| 312 |
"step": 2400
|
| 313 |
},
|
| 314 |
{
|
| 315 |
"epoch": 10.0,
|
| 316 |
-
"grad_norm": 1.1837812662124634,
|
| 317 |
"learning_rate": 0.0005,
|
| 318 |
-
"loss": 0.
|
| 319 |
"step": 2500
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 10.0,
|
| 323 |
-
"eval_accuracy": 0.
|
| 324 |
-
"eval_loss":
|
| 325 |
-
"eval_runtime":
|
| 326 |
-
"eval_samples_per_second":
|
| 327 |
-
"eval_steps_per_second":
|
| 328 |
"step": 2500
|
| 329 |
},
|
| 330 |
{
|
| 331 |
"epoch": 10.0,
|
| 332 |
-
"eval_exact_match":
|
| 333 |
-
"eval_f1":
|
| 334 |
"step": 2500
|
| 335 |
},
|
| 336 |
{
|
| 337 |
"epoch": 10.4,
|
| 338 |
-
"grad_norm": 1.1431084871292114,
|
| 339 |
"learning_rate": 0.0005,
|
| 340 |
-
"loss": 0.
|
| 341 |
"step": 2600
|
| 342 |
},
|
| 343 |
{
|
| 344 |
"epoch": 10.8,
|
| 345 |
-
"grad_norm": 1.0149098634719849,
|
| 346 |
"learning_rate": 0.0005,
|
| 347 |
-
"loss": 0.
|
| 348 |
"step": 2700
|
| 349 |
},
|
| 350 |
{
|
| 351 |
"epoch": 11.0,
|
| 352 |
-
"eval_accuracy": 0.
|
| 353 |
-
"eval_loss":
|
| 354 |
-
"eval_runtime": 5.
|
| 355 |
-
"eval_samples_per_second":
|
| 356 |
-
"eval_steps_per_second":
|
| 357 |
"step": 2750
|
| 358 |
},
|
| 359 |
{
|
| 360 |
"epoch": 11.0,
|
| 361 |
-
"eval_exact_match":
|
| 362 |
-
"eval_f1":
|
| 363 |
"step": 2750
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 11.2,
|
| 367 |
-
"grad_norm": 0.666008710861206,
|
| 368 |
"learning_rate": 0.0005,
|
| 369 |
-
"loss": 0.
|
| 370 |
"step": 2800
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 11.6,
|
| 374 |
-
"grad_norm": 0.8676968812942505,
|
| 375 |
"learning_rate": 0.0005,
|
| 376 |
-
"loss": 0.
|
| 377 |
"step": 2900
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 12.0,
|
| 381 |
-
"grad_norm": 0.8371489644050598,
|
| 382 |
"learning_rate": 0.0005,
|
| 383 |
-
"loss": 0.
|
| 384 |
"step": 3000
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 12.0,
|
| 388 |
-
"eval_accuracy": 0.
|
| 389 |
-
"eval_loss":
|
| 390 |
-
"eval_runtime":
|
| 391 |
-
"eval_samples_per_second":
|
| 392 |
-
"eval_steps_per_second":
|
| 393 |
"step": 3000
|
| 394 |
},
|
| 395 |
{
|
| 396 |
"epoch": 12.0,
|
| 397 |
-
"eval_exact_match":
|
| 398 |
-
"eval_f1":
|
| 399 |
"step": 3000
|
| 400 |
},
|
| 401 |
{
|
| 402 |
"epoch": 12.4,
|
| 403 |
-
"grad_norm": 0.8826413750648499,
|
| 404 |
"learning_rate": 0.0005,
|
| 405 |
-
"loss": 0.
|
| 406 |
"step": 3100
|
| 407 |
},
|
| 408 |
{
|
| 409 |
"epoch": 12.8,
|
| 410 |
-
"grad_norm": 0.9569880962371826,
|
| 411 |
"learning_rate": 0.0005,
|
| 412 |
-
"loss": 0.
|
| 413 |
"step": 3200
|
| 414 |
},
|
| 415 |
{
|
| 416 |
"epoch": 13.0,
|
| 417 |
-
"eval_accuracy": 0.
|
| 418 |
-
"eval_loss":
|
| 419 |
-
"eval_runtime":
|
| 420 |
-
"eval_samples_per_second":
|
| 421 |
-
"eval_steps_per_second":
|
| 422 |
"step": 3250
|
| 423 |
},
|
| 424 |
{
|
| 425 |
"epoch": 13.0,
|
| 426 |
-
"eval_exact_match":
|
| 427 |
-
"eval_f1":
|
| 428 |
"step": 3250
|
| 429 |
},
|
| 430 |
{
|
| 431 |
"epoch": 13.2,
|
| 432 |
-
"grad_norm": 0.8115692734718323,
|
| 433 |
"learning_rate": 0.0005,
|
| 434 |
-
"loss": 0.
|
| 435 |
"step": 3300
|
| 436 |
},
|
| 437 |
{
|
| 438 |
"epoch": 13.6,
|
| 439 |
-
"grad_norm": 0.9201633334159851,
|
| 440 |
"learning_rate": 0.0005,
|
| 441 |
-
"loss": 0.
|
| 442 |
"step": 3400
|
| 443 |
},
|
| 444 |
{
|
| 445 |
"epoch": 14.0,
|
| 446 |
-
"grad_norm": 1.0557641983032227,
|
| 447 |
"learning_rate": 0.0005,
|
| 448 |
-
"loss": 0.
|
| 449 |
"step": 3500
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 14.0,
|
| 453 |
-
"eval_accuracy": 0.
|
| 454 |
-
"eval_loss":
|
| 455 |
-
"eval_runtime":
|
| 456 |
-
"eval_samples_per_second":
|
| 457 |
-
"eval_steps_per_second":
|
| 458 |
"step": 3500
|
| 459 |
},
|
| 460 |
{
|
| 461 |
"epoch": 14.0,
|
| 462 |
-
"eval_exact_match":
|
| 463 |
-
"eval_f1":
|
| 464 |
"step": 3500
|
| 465 |
},
|
| 466 |
{
|
| 467 |
"epoch": 14.4,
|
| 468 |
-
"grad_norm": 0.7237015962600708,
|
| 469 |
"learning_rate": 0.0005,
|
| 470 |
-
"loss": 0.
|
| 471 |
"step": 3600
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 14.8,
|
| 475 |
-
"grad_norm": 0.8207970857620239,
|
| 476 |
"learning_rate": 0.0005,
|
| 477 |
-
"loss": 0.
|
| 478 |
"step": 3700
|
| 479 |
},
|
| 480 |
{
|
| 481 |
"epoch": 15.0,
|
| 482 |
-
"eval_accuracy": 0.
|
| 483 |
-
"eval_loss":
|
| 484 |
-
"eval_runtime":
|
| 485 |
-
"eval_samples_per_second":
|
| 486 |
-
"eval_steps_per_second":
|
| 487 |
"step": 3750
|
| 488 |
},
|
| 489 |
{
|
| 490 |
"epoch": 15.0,
|
| 491 |
-
"eval_exact_match":
|
| 492 |
-
"eval_f1":
|
| 493 |
"step": 3750
|
| 494 |
},
|
| 495 |
{
|
| 496 |
"epoch": 15.2,
|
| 497 |
-
"grad_norm": 0.9946625232696533,
|
| 498 |
"learning_rate": 0.0005,
|
| 499 |
-
"loss": 0.
|
| 500 |
"step": 3800
|
| 501 |
},
|
| 502 |
{
|
| 503 |
"epoch": 15.6,
|
| 504 |
-
"grad_norm": 1.109531283378601,
|
| 505 |
"learning_rate": 0.0005,
|
| 506 |
-
"loss": 0.
|
| 507 |
"step": 3900
|
| 508 |
},
|
| 509 |
{
|
| 510 |
"epoch": 16.0,
|
| 511 |
-
"grad_norm": 0.8117135167121887,
|
| 512 |
"learning_rate": 0.0005,
|
| 513 |
-
"loss": 0.
|
| 514 |
"step": 4000
|
| 515 |
},
|
| 516 |
{
|
| 517 |
"epoch": 16.0,
|
| 518 |
-
"eval_accuracy": 0.
|
| 519 |
-
"eval_loss":
|
| 520 |
-
"eval_runtime":
|
| 521 |
-
"eval_samples_per_second":
|
| 522 |
-
"eval_steps_per_second":
|
| 523 |
"step": 4000
|
| 524 |
},
|
| 525 |
{
|
| 526 |
"epoch": 16.0,
|
| 527 |
-
"eval_exact_match":
|
| 528 |
-
"eval_f1":
|
| 529 |
"step": 4000
|
| 530 |
},
|
| 531 |
{
|
| 532 |
"epoch": 16.4,
|
| 533 |
-
"grad_norm": 0.8689247369766235,
|
| 534 |
"learning_rate": 0.0005,
|
| 535 |
-
"loss": 0.
|
| 536 |
"step": 4100
|
| 537 |
},
|
| 538 |
{
|
| 539 |
"epoch": 16.8,
|
| 540 |
-
"grad_norm": 0.8414099812507629,
|
| 541 |
"learning_rate": 0.0005,
|
| 542 |
-
"loss": 0.
|
| 543 |
"step": 4200
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 17.0,
|
| 547 |
-
"eval_accuracy": 0.
|
| 548 |
-
"eval_loss":
|
| 549 |
-
"eval_runtime": 5.
|
| 550 |
-
"eval_samples_per_second":
|
| 551 |
-
"eval_steps_per_second":
|
| 552 |
"step": 4250
|
| 553 |
},
|
| 554 |
{
|
| 555 |
"epoch": 17.0,
|
| 556 |
-
"eval_exact_match":
|
| 557 |
-
"eval_f1":
|
| 558 |
"step": 4250
|
| 559 |
},
|
| 560 |
{
|
| 561 |
"epoch": 17.2,
|
| 562 |
-
"grad_norm": 0.8533599972724915,
|
| 563 |
"learning_rate": 0.0005,
|
| 564 |
-
"loss": 0.
|
| 565 |
"step": 4300
|
| 566 |
},
|
| 567 |
{
|
| 568 |
"epoch": 17.6,
|
| 569 |
-
"grad_norm": 1.02438485622406,
|
| 570 |
"learning_rate": 0.0005,
|
| 571 |
-
"loss": 0.
|
| 572 |
"step": 4400
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 18.0,
|
| 576 |
-
"grad_norm": 1.0211039781570435,
|
| 577 |
"learning_rate": 0.0005,
|
| 578 |
-
"loss": 0.
|
| 579 |
"step": 4500
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 18.0,
|
| 583 |
-
"eval_accuracy": 0.
|
| 584 |
-
"eval_loss":
|
| 585 |
-
"eval_runtime":
|
| 586 |
-
"eval_samples_per_second":
|
| 587 |
-
"eval_steps_per_second":
|
| 588 |
"step": 4500
|
| 589 |
},
|
| 590 |
{
|
| 591 |
"epoch": 18.0,
|
| 592 |
-
"eval_exact_match":
|
| 593 |
-
"eval_f1":
|
| 594 |
"step": 4500
|
| 595 |
},
|
| 596 |
{
|
| 597 |
"epoch": 18.4,
|
| 598 |
-
"grad_norm": 0.692489743232727,
|
| 599 |
"learning_rate": 0.0005,
|
| 600 |
-
"loss": 0.
|
| 601 |
"step": 4600
|
| 602 |
},
|
| 603 |
{
|
| 604 |
"epoch": 18.8,
|
| 605 |
-
"grad_norm": 1.098289966583252,
|
| 606 |
"learning_rate": 0.0005,
|
| 607 |
-
"loss": 0.
|
| 608 |
"step": 4700
|
| 609 |
},
|
| 610 |
{
|
| 611 |
"epoch": 19.0,
|
| 612 |
-
"eval_accuracy": 0.
|
| 613 |
-
"eval_loss":
|
| 614 |
-
"eval_runtime":
|
| 615 |
-
"eval_samples_per_second":
|
| 616 |
-
"eval_steps_per_second":
|
| 617 |
"step": 4750
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 19.0,
|
| 621 |
-
"eval_exact_match":
|
| 622 |
-
"eval_f1":
|
| 623 |
"step": 4750
|
| 624 |
},
|
| 625 |
{
|
| 626 |
"epoch": 19.2,
|
| 627 |
-
"grad_norm": 0.7502797842025757,
|
| 628 |
"learning_rate": 0.0005,
|
| 629 |
-
"loss": 0.
|
| 630 |
"step": 4800
|
| 631 |
},
|
| 632 |
{
|
| 633 |
"epoch": 19.6,
|
| 634 |
-
"grad_norm": 1.0003832578659058,
|
| 635 |
"learning_rate": 0.0005,
|
| 636 |
-
"loss": 0.
|
| 637 |
"step": 4900
|
| 638 |
},
|
| 639 |
{
|
| 640 |
"epoch": 20.0,
|
| 641 |
-
"grad_norm": 0.9703432321548462,
|
| 642 |
"learning_rate": 0.0005,
|
| 643 |
-
"loss": 0.
|
| 644 |
"step": 5000
|
| 645 |
},
|
| 646 |
{
|
| 647 |
"epoch": 20.0,
|
| 648 |
-
"eval_accuracy": 0.
|
| 649 |
-
"eval_loss":
|
| 650 |
-
"eval_runtime": 5.
|
| 651 |
-
"eval_samples_per_second":
|
| 652 |
-
"eval_steps_per_second":
|
| 653 |
"step": 5000
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 20.0,
|
| 657 |
-
"eval_exact_match":
|
| 658 |
-
"eval_f1":
|
| 659 |
"step": 5000
|
| 660 |
},
|
| 661 |
{
|
| 662 |
"epoch": 20.4,
|
| 663 |
-
"grad_norm": 0.8916403651237488,
|
| 664 |
"learning_rate": 0.0005,
|
| 665 |
-
"loss": 0.
|
| 666 |
"step": 5100
|
| 667 |
},
|
| 668 |
{
|
| 669 |
"epoch": 20.8,
|
| 670 |
-
"grad_norm": 0.8316965103149414,
|
| 671 |
"learning_rate": 0.0005,
|
| 672 |
-
"loss": 0.
|
| 673 |
"step": 5200
|
| 674 |
},
|
| 675 |
{
|
| 676 |
"epoch": 21.0,
|
| 677 |
-
"eval_accuracy": 0.
|
| 678 |
-
"eval_loss": 3.
|
| 679 |
-
"eval_runtime":
|
| 680 |
-
"eval_samples_per_second":
|
| 681 |
-
"eval_steps_per_second":
|
| 682 |
"step": 5250
|
| 683 |
},
|
| 684 |
{
|
| 685 |
"epoch": 21.0,
|
| 686 |
-
"eval_exact_match":
|
| 687 |
-
"eval_f1":
|
| 688 |
"step": 5250
|
| 689 |
}
|
| 690 |
],
|
| 691 |
"logging_steps": 100,
|
| 692 |
"max_steps": 12500,
|
| 693 |
-
"num_input_tokens_seen": 0,
|
| 694 |
"num_train_epochs": 50,
|
| 695 |
"save_steps": 500,
|
| 696 |
-
"
|
| 697 |
-
"TrainerControl": {
|
| 698 |
-
"args": {
|
| 699 |
-
"should_epoch_stop": false,
|
| 700 |
-
"should_evaluate": false,
|
| 701 |
-
"should_log": false,
|
| 702 |
-
"should_save": true,
|
| 703 |
-
"should_training_stop": false
|
| 704 |
-
},
|
| 705 |
-
"attributes": {}
|
| 706 |
-
}
|
| 707 |
-
},
|
| 708 |
-
"total_flos": 1.606497816417403e+17,
|
| 709 |
-
"train_batch_size": 2,
|
| 710 |
"trial_name": null,
|
| 711 |
"trial_params": null
|
| 712 |
}
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.4,
|
|
|
|
| 13 |
"learning_rate": 0.0005,
|
| 14 |
+
"loss": 2.3712,
|
| 15 |
"step": 100
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.8,
|
|
|
|
| 19 |
"learning_rate": 0.0005,
|
| 20 |
+
"loss": 1.8732,
|
| 21 |
"step": 200
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 1.0,
|
| 25 |
+
"eval_accuracy": 0.593873417721519,
|
| 26 |
+
"eval_loss": 2.011091947555542,
|
| 27 |
+
"eval_runtime": 4.861,
|
| 28 |
+
"eval_samples_per_second": 102.859,
|
| 29 |
+
"eval_steps_per_second": 12.96,
|
| 30 |
"step": 250
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 1.0,
|
| 34 |
+
"eval_exact_match": 0.0,
|
| 35 |
+
"eval_f1": 0.32835692091264185,
|
| 36 |
"step": 250
|
| 37 |
},
|
| 38 |
{
|
| 39 |
"epoch": 1.2,
|
|
|
|
| 40 |
"learning_rate": 0.0005,
|
| 41 |
+
"loss": 1.7329,
|
| 42 |
"step": 300
|
| 43 |
},
|
| 44 |
{
|
| 45 |
"epoch": 1.6,
|
|
|
|
| 46 |
"learning_rate": 0.0005,
|
| 47 |
+
"loss": 1.6233,
|
| 48 |
"step": 400
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 2.0,
|
|
|
|
| 52 |
"learning_rate": 0.0005,
|
| 53 |
+
"loss": 1.6142,
|
| 54 |
"step": 500
|
| 55 |
},
|
| 56 |
{
|
| 57 |
"epoch": 2.0,
|
| 58 |
+
"eval_accuracy": 0.6051139240506329,
|
| 59 |
+
"eval_loss": 1.844284176826477,
|
| 60 |
+
"eval_runtime": 4.8205,
|
| 61 |
+
"eval_samples_per_second": 103.724,
|
| 62 |
+
"eval_steps_per_second": 13.069,
|
| 63 |
"step": 500
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 2.0,
|
| 67 |
+
"eval_exact_match": 0.0,
|
| 68 |
+
"eval_f1": 0.7215873015873017,
|
| 69 |
"step": 500
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.4,
|
|
|
|
| 73 |
"learning_rate": 0.0005,
|
| 74 |
+
"loss": 1.1625,
|
| 75 |
"step": 600
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 2.8,
|
|
|
|
| 79 |
"learning_rate": 0.0005,
|
| 80 |
+
"loss": 1.206,
|
| 81 |
"step": 700
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 3.0,
|
| 85 |
+
"eval_accuracy": 0.6007341772151898,
|
| 86 |
+
"eval_loss": 1.981817603111267,
|
| 87 |
+
"eval_runtime": 4.671,
|
| 88 |
+
"eval_samples_per_second": 107.044,
|
| 89 |
+
"eval_steps_per_second": 13.487,
|
| 90 |
"step": 750
|
| 91 |
},
|
| 92 |
{
|
| 93 |
"epoch": 3.0,
|
| 94 |
+
"eval_exact_match": 0.0,
|
| 95 |
+
"eval_f1": 0.4163636363636364,
|
| 96 |
"step": 750
|
| 97 |
},
|
| 98 |
{
|
| 99 |
"epoch": 3.2,
|
|
|
|
| 100 |
"learning_rate": 0.0005,
|
| 101 |
+
"loss": 1.0064,
|
| 102 |
"step": 800
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 3.6,
|
|
|
|
| 106 |
"learning_rate": 0.0005,
|
| 107 |
+
"loss": 0.8141,
|
| 108 |
"step": 900
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 4.0,
|
|
|
|
| 112 |
"learning_rate": 0.0005,
|
| 113 |
+
"loss": 0.8693,
|
| 114 |
"step": 1000
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 4.0,
|
| 118 |
+
"eval_accuracy": 0.594126582278481,
|
| 119 |
+
"eval_loss": 2.2100281715393066,
|
| 120 |
+
"eval_runtime": 5.2488,
|
| 121 |
+
"eval_samples_per_second": 95.26,
|
| 122 |
+
"eval_steps_per_second": 12.003,
|
| 123 |
"step": 1000
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 4.0,
|
| 127 |
+
"eval_exact_match": 0.2,
|
| 128 |
+
"eval_f1": 1.559097642686839,
|
| 129 |
"step": 1000
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 4.4,
|
|
|
|
| 133 |
"learning_rate": 0.0005,
|
| 134 |
+
"loss": 0.5477,
|
| 135 |
"step": 1100
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 4.8,
|
|
|
|
| 139 |
"learning_rate": 0.0005,
|
| 140 |
+
"loss": 0.6023,
|
| 141 |
"step": 1200
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"epoch": 5.0,
|
| 145 |
+
"eval_accuracy": 0.5910126582278481,
|
| 146 |
+
"eval_loss": 2.375615119934082,
|
| 147 |
+
"eval_runtime": 4.4156,
|
| 148 |
+
"eval_samples_per_second": 113.235,
|
| 149 |
+
"eval_steps_per_second": 14.268,
|
| 150 |
"step": 1250
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 5.0,
|
| 154 |
+
"eval_exact_match": 0.4,
|
| 155 |
+
"eval_f1": 2.41971862073413,
|
| 156 |
"step": 1250
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 5.2,
|
|
|
|
| 160 |
"learning_rate": 0.0005,
|
| 161 |
+
"loss": 0.514,
|
| 162 |
"step": 1300
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 5.6,
|
|
|
|
| 166 |
"learning_rate": 0.0005,
|
| 167 |
+
"loss": 0.4479,
|
| 168 |
"step": 1400
|
| 169 |
},
|
| 170 |
{
|
| 171 |
"epoch": 6.0,
|
|
|
|
| 172 |
"learning_rate": 0.0005,
|
| 173 |
+
"loss": 0.4717,
|
| 174 |
"step": 1500
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 6.0,
|
| 178 |
+
"eval_accuracy": 0.5895696202531645,
|
| 179 |
+
"eval_loss": 2.542104482650757,
|
| 180 |
+
"eval_runtime": 5.1447,
|
| 181 |
+
"eval_samples_per_second": 97.188,
|
| 182 |
+
"eval_steps_per_second": 12.246,
|
| 183 |
"step": 1500
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 6.0,
|
| 187 |
+
"eval_exact_match": 9.8,
|
| 188 |
+
"eval_f1": 17.19683736589267,
|
| 189 |
"step": 1500
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 6.4,
|
|
|
|
| 193 |
"learning_rate": 0.0005,
|
| 194 |
+
"loss": 0.3723,
|
| 195 |
"step": 1600
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 6.8,
|
|
|
|
| 199 |
"learning_rate": 0.0005,
|
| 200 |
+
"loss": 0.3938,
|
| 201 |
"step": 1700
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 7.0,
|
| 205 |
+
"eval_accuracy": 0.5891139240506329,
|
| 206 |
+
"eval_loss": 2.658656597137451,
|
| 207 |
+
"eval_runtime": 4.7175,
|
| 208 |
+
"eval_samples_per_second": 105.987,
|
| 209 |
+
"eval_steps_per_second": 13.354,
|
| 210 |
"step": 1750
|
| 211 |
},
|
| 212 |
{
|
| 213 |
"epoch": 7.0,
|
| 214 |
+
"eval_exact_match": 3.6,
|
| 215 |
+
"eval_f1": 8.743843476498272,
|
| 216 |
"step": 1750
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 7.2,
|
|
|
|
| 220 |
"learning_rate": 0.0005,
|
| 221 |
+
"loss": 0.3727,
|
| 222 |
"step": 1800
|
| 223 |
},
|
| 224 |
{
|
| 225 |
"epoch": 7.6,
|
|
|
|
| 226 |
"learning_rate": 0.0005,
|
| 227 |
+
"loss": 0.3538,
|
| 228 |
"step": 1900
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 8.0,
|
|
|
|
| 232 |
"learning_rate": 0.0005,
|
| 233 |
+
"loss": 0.3697,
|
| 234 |
"step": 2000
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 8.0,
|
| 238 |
+
"eval_accuracy": 0.5872911392405064,
|
| 239 |
+
"eval_loss": 2.7531518936157227,
|
| 240 |
+
"eval_runtime": 4.7213,
|
| 241 |
+
"eval_samples_per_second": 105.902,
|
| 242 |
+
"eval_steps_per_second": 13.344,
|
| 243 |
"step": 2000
|
| 244 |
},
|
| 245 |
{
|
| 246 |
"epoch": 8.0,
|
| 247 |
+
"eval_exact_match": 0.0,
|
| 248 |
+
"eval_f1": 0.3733369257948794,
|
| 249 |
"step": 2000
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 8.4,
|
|
|
|
| 253 |
"learning_rate": 0.0005,
|
| 254 |
+
"loss": 0.3431,
|
| 255 |
"step": 2100
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 8.8,
|
|
|
|
| 259 |
"learning_rate": 0.0005,
|
| 260 |
+
"loss": 0.3617,
|
| 261 |
"step": 2200
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 9.0,
|
| 265 |
+
"eval_accuracy": 0.5870126582278481,
|
| 266 |
+
"eval_loss": 2.7664403915405273,
|
| 267 |
+
"eval_runtime": 4.7203,
|
| 268 |
+
"eval_samples_per_second": 105.925,
|
| 269 |
+
"eval_steps_per_second": 13.347,
|
| 270 |
"step": 2250
|
| 271 |
},
|
| 272 |
{
|
| 273 |
"epoch": 9.0,
|
| 274 |
+
"eval_exact_match": 0.0,
|
| 275 |
+
"eval_f1": 0.18,
|
| 276 |
"step": 2250
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 9.2,
|
|
|
|
| 280 |
"learning_rate": 0.0005,
|
| 281 |
+
"loss": 0.3424,
|
| 282 |
"step": 2300
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 9.6,
|
|
|
|
| 286 |
"learning_rate": 0.0005,
|
| 287 |
+
"loss": 0.3418,
|
| 288 |
"step": 2400
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 10.0,
|
|
|
|
| 292 |
"learning_rate": 0.0005,
|
| 293 |
+
"loss": 0.3607,
|
| 294 |
"step": 2500
|
| 295 |
},
|
| 296 |
{
|
| 297 |
"epoch": 10.0,
|
| 298 |
+
"eval_accuracy": 0.5866835443037974,
|
| 299 |
+
"eval_loss": 2.8513779640197754,
|
| 300 |
+
"eval_runtime": 4.5658,
|
| 301 |
+
"eval_samples_per_second": 109.511,
|
| 302 |
+
"eval_steps_per_second": 13.798,
|
| 303 |
"step": 2500
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 10.0,
|
| 307 |
+
"eval_exact_match": 15.8,
|
| 308 |
+
"eval_f1": 24.03880563880564,
|
| 309 |
"step": 2500
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 10.4,
|
|
|
|
| 313 |
"learning_rate": 0.0005,
|
| 314 |
+
"loss": 0.339,
|
| 315 |
"step": 2600
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"epoch": 10.8,
|
|
|
|
| 319 |
"learning_rate": 0.0005,
|
| 320 |
+
"loss": 0.3414,
|
| 321 |
"step": 2700
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 11.0,
|
| 325 |
+
"eval_accuracy": 0.5860759493670886,
|
| 326 |
+
"eval_loss": 2.8931641578674316,
|
| 327 |
+
"eval_runtime": 5.1473,
|
| 328 |
+
"eval_samples_per_second": 97.139,
|
| 329 |
+
"eval_steps_per_second": 12.24,
|
| 330 |
"step": 2750
|
| 331 |
},
|
| 332 |
{
|
| 333 |
"epoch": 11.0,
|
| 334 |
+
"eval_exact_match": 0.0,
|
| 335 |
+
"eval_f1": 1.8004313814840132,
|
| 336 |
"step": 2750
|
| 337 |
},
|
| 338 |
{
|
| 339 |
"epoch": 11.2,
|
|
|
|
| 340 |
"learning_rate": 0.0005,
|
| 341 |
+
"loss": 0.3302,
|
| 342 |
"step": 2800
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 11.6,
|
|
|
|
| 346 |
"learning_rate": 0.0005,
|
| 347 |
+
"loss": 0.329,
|
| 348 |
"step": 2900
|
| 349 |
},
|
| 350 |
{
|
| 351 |
"epoch": 12.0,
|
|
|
|
| 352 |
"learning_rate": 0.0005,
|
| 353 |
+
"loss": 0.3439,
|
| 354 |
"step": 3000
|
| 355 |
},
|
| 356 |
{
|
| 357 |
"epoch": 12.0,
|
| 358 |
+
"eval_accuracy": 0.5854683544303797,
|
| 359 |
+
"eval_loss": 2.9544754028320312,
|
| 360 |
+
"eval_runtime": 4.5673,
|
| 361 |
+
"eval_samples_per_second": 109.473,
|
| 362 |
+
"eval_steps_per_second": 13.794,
|
| 363 |
"step": 3000
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 12.0,
|
| 367 |
+
"eval_exact_match": 0.8,
|
| 368 |
+
"eval_f1": 3.4374976657329603,
|
| 369 |
"step": 3000
|
| 370 |
},
|
| 371 |
{
|
| 372 |
"epoch": 12.4,
|
|
|
|
| 373 |
"learning_rate": 0.0005,
|
| 374 |
+
"loss": 0.3182,
|
| 375 |
"step": 3100
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 12.8,
|
|
|
|
| 379 |
"learning_rate": 0.0005,
|
| 380 |
+
"loss": 0.335,
|
| 381 |
"step": 3200
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 13.0,
|
| 385 |
+
"eval_accuracy": 0.5842784810126582,
|
| 386 |
+
"eval_loss": 2.899094343185425,
|
| 387 |
+
"eval_runtime": 4.7188,
|
| 388 |
+
"eval_samples_per_second": 105.96,
|
| 389 |
+
"eval_steps_per_second": 13.351,
|
| 390 |
"step": 3250
|
| 391 |
},
|
| 392 |
{
|
| 393 |
"epoch": 13.0,
|
| 394 |
+
"eval_exact_match": 14.4,
|
| 395 |
+
"eval_f1": 23.30256743256743,
|
| 396 |
"step": 3250
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 13.2,
|
|
|
|
| 400 |
"learning_rate": 0.0005,
|
| 401 |
+
"loss": 0.327,
|
| 402 |
"step": 3300
|
| 403 |
},
|
| 404 |
{
|
| 405 |
"epoch": 13.6,
|
|
|
|
| 406 |
"learning_rate": 0.0005,
|
| 407 |
+
"loss": 0.3223,
|
| 408 |
"step": 3400
|
| 409 |
},
|
| 410 |
{
|
| 411 |
"epoch": 14.0,
|
|
|
|
| 412 |
"learning_rate": 0.0005,
|
| 413 |
+
"loss": 0.3391,
|
| 414 |
"step": 3500
|
| 415 |
},
|
| 416 |
{
|
| 417 |
"epoch": 14.0,
|
| 418 |
+
"eval_accuracy": 0.5839746835443038,
|
| 419 |
+
"eval_loss": 2.879316806793213,
|
| 420 |
+
"eval_runtime": 4.415,
|
| 421 |
+
"eval_samples_per_second": 113.251,
|
| 422 |
+
"eval_steps_per_second": 14.27,
|
| 423 |
"step": 3500
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"epoch": 14.0,
|
| 427 |
+
"eval_exact_match": 17.2,
|
| 428 |
+
"eval_f1": 25.861362134597435,
|
| 429 |
"step": 3500
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 14.4,
|
|
|
|
| 433 |
"learning_rate": 0.0005,
|
| 434 |
+
"loss": 0.3128,
|
| 435 |
"step": 3600
|
| 436 |
},
|
| 437 |
{
|
| 438 |
"epoch": 14.8,
|
|
|
|
| 439 |
"learning_rate": 0.0005,
|
| 440 |
+
"loss": 0.328,
|
| 441 |
"step": 3700
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 15.0,
|
| 445 |
+
"eval_accuracy": 0.5851392405063292,
|
| 446 |
+
"eval_loss": 2.8953680992126465,
|
| 447 |
+
"eval_runtime": 4.7507,
|
| 448 |
+
"eval_samples_per_second": 105.249,
|
| 449 |
+
"eval_steps_per_second": 13.261,
|
| 450 |
"step": 3750
|
| 451 |
},
|
| 452 |
{
|
| 453 |
"epoch": 15.0,
|
| 454 |
+
"eval_exact_match": 15.8,
|
| 455 |
+
"eval_f1": 23.802868797868808,
|
| 456 |
"step": 3750
|
| 457 |
},
|
| 458 |
{
|
| 459 |
"epoch": 15.2,
|
|
|
|
| 460 |
"learning_rate": 0.0005,
|
| 461 |
+
"loss": 0.3233,
|
| 462 |
"step": 3800
|
| 463 |
},
|
| 464 |
{
|
| 465 |
"epoch": 15.6,
|
|
|
|
| 466 |
"learning_rate": 0.0005,
|
| 467 |
+
"loss": 0.3216,
|
| 468 |
"step": 3900
|
| 469 |
},
|
| 470 |
{
|
| 471 |
"epoch": 16.0,
|
|
|
|
| 472 |
"learning_rate": 0.0005,
|
| 473 |
+
"loss": 0.3351,
|
| 474 |
"step": 4000
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 16.0,
|
| 478 |
+
"eval_accuracy": 0.5838481012658228,
|
| 479 |
+
"eval_loss": 2.913999080657959,
|
| 480 |
+
"eval_runtime": 4.5211,
|
| 481 |
+
"eval_samples_per_second": 110.593,
|
| 482 |
+
"eval_steps_per_second": 13.935,
|
| 483 |
"step": 4000
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 16.0,
|
| 487 |
+
"eval_exact_match": 15.8,
|
| 488 |
+
"eval_f1": 23.843056551291856,
|
| 489 |
"step": 4000
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 16.4,
|
|
|
|
| 493 |
"learning_rate": 0.0005,
|
| 494 |
+
"loss": 0.3087,
|
| 495 |
"step": 4100
|
| 496 |
},
|
| 497 |
{
|
| 498 |
"epoch": 16.8,
|
|
|
|
| 499 |
"learning_rate": 0.0005,
|
| 500 |
+
"loss": 0.3252,
|
| 501 |
"step": 4200
|
| 502 |
},
|
| 503 |
{
|
| 504 |
"epoch": 17.0,
|
| 505 |
+
"eval_accuracy": 0.5825063291139241,
|
| 506 |
+
"eval_loss": 2.929701089859009,
|
| 507 |
+
"eval_runtime": 5.1507,
|
| 508 |
+
"eval_samples_per_second": 97.074,
|
| 509 |
+
"eval_steps_per_second": 12.231,
|
| 510 |
"step": 4250
|
| 511 |
},
|
| 512 |
{
|
| 513 |
"epoch": 17.0,
|
| 514 |
+
"eval_exact_match": 17.6,
|
| 515 |
+
"eval_f1": 25.533542568542583,
|
| 516 |
"step": 4250
|
| 517 |
},
|
| 518 |
{
|
| 519 |
"epoch": 17.2,
|
|
|
|
| 520 |
"learning_rate": 0.0005,
|
| 521 |
+
"loss": 0.3152,
|
| 522 |
"step": 4300
|
| 523 |
},
|
| 524 |
{
|
| 525 |
"epoch": 17.6,
|
|
|
|
| 526 |
"learning_rate": 0.0005,
|
| 527 |
+
"loss": 0.3161,
|
| 528 |
"step": 4400
|
| 529 |
},
|
| 530 |
{
|
| 531 |
"epoch": 18.0,
|
|
|
|
| 532 |
"learning_rate": 0.0005,
|
| 533 |
+
"loss": 0.332,
|
| 534 |
"step": 4500
|
| 535 |
},
|
| 536 |
{
|
| 537 |
"epoch": 18.0,
|
| 538 |
+
"eval_accuracy": 0.5833924050632912,
|
| 539 |
+
"eval_loss": 2.9811997413635254,
|
| 540 |
+
"eval_runtime": 4.5707,
|
| 541 |
+
"eval_samples_per_second": 109.392,
|
| 542 |
+
"eval_steps_per_second": 13.783,
|
| 543 |
"step": 4500
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 18.0,
|
| 547 |
+
"eval_exact_match": 14.8,
|
| 548 |
+
"eval_f1": 23.04023215346745,
|
| 549 |
"step": 4500
|
| 550 |
},
|
| 551 |
{
|
| 552 |
"epoch": 18.4,
|
|
|
|
| 553 |
"learning_rate": 0.0005,
|
| 554 |
+
"loss": 0.3089,
|
| 555 |
"step": 4600
|
| 556 |
},
|
| 557 |
{
|
| 558 |
"epoch": 18.8,
|
|
|
|
| 559 |
"learning_rate": 0.0005,
|
| 560 |
+
"loss": 0.324,
|
| 561 |
"step": 4700
|
| 562 |
},
|
| 563 |
{
|
| 564 |
"epoch": 19.0,
|
| 565 |
+
"eval_accuracy": 0.5807848101265822,
|
| 566 |
+
"eval_loss": 2.982343912124634,
|
| 567 |
+
"eval_runtime": 4.5156,
|
| 568 |
+
"eval_samples_per_second": 110.728,
|
| 569 |
+
"eval_steps_per_second": 13.952,
|
| 570 |
"step": 4750
|
| 571 |
},
|
| 572 |
{
|
| 573 |
"epoch": 19.0,
|
| 574 |
+
"eval_exact_match": 17.2,
|
| 575 |
+
"eval_f1": 24.75057253204313,
|
| 576 |
"step": 4750
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 19.2,
|
|
|
|
| 580 |
"learning_rate": 0.0005,
|
| 581 |
+
"loss": 0.3169,
|
| 582 |
"step": 4800
|
| 583 |
},
|
| 584 |
{
|
| 585 |
"epoch": 19.6,
|
|
|
|
| 586 |
"learning_rate": 0.0005,
|
| 587 |
+
"loss": 0.3116,
|
| 588 |
"step": 4900
|
| 589 |
},
|
| 590 |
{
|
| 591 |
"epoch": 20.0,
|
|
|
|
| 592 |
"learning_rate": 0.0005,
|
| 593 |
+
"loss": 0.3329,
|
| 594 |
"step": 5000
|
| 595 |
},
|
| 596 |
{
|
| 597 |
"epoch": 20.0,
|
| 598 |
+
"eval_accuracy": 0.5813164556962025,
|
| 599 |
+
"eval_loss": 2.9419918060302734,
|
| 600 |
+
"eval_runtime": 5.1584,
|
| 601 |
+
"eval_samples_per_second": 96.929,
|
| 602 |
+
"eval_steps_per_second": 12.213,
|
| 603 |
"step": 5000
|
| 604 |
},
|
| 605 |
{
|
| 606 |
"epoch": 20.0,
|
| 607 |
+
"eval_exact_match": 17.0,
|
| 608 |
+
"eval_f1": 24.681692914928213,
|
| 609 |
"step": 5000
|
| 610 |
},
|
| 611 |
{
|
| 612 |
"epoch": 20.4,
|
|
|
|
| 613 |
"learning_rate": 0.0005,
|
| 614 |
+
"loss": 0.3052,
|
| 615 |
"step": 5100
|
| 616 |
},
|
| 617 |
{
|
| 618 |
"epoch": 20.8,
|
|
|
|
| 619 |
"learning_rate": 0.0005,
|
| 620 |
+
"loss": 0.3242,
|
| 621 |
"step": 5200
|
| 622 |
},
|
| 623 |
{
|
| 624 |
"epoch": 21.0,
|
| 625 |
+
"eval_accuracy": 0.5806075949367089,
|
| 626 |
+
"eval_loss": 3.0366506576538086,
|
| 627 |
+
"eval_runtime": 4.4135,
|
| 628 |
+
"eval_samples_per_second": 113.288,
|
| 629 |
+
"eval_steps_per_second": 14.274,
|
| 630 |
"step": 5250
|
| 631 |
},
|
| 632 |
{
|
| 633 |
"epoch": 21.0,
|
| 634 |
+
"eval_exact_match": 15.8,
|
| 635 |
+
"eval_f1": 24.507380005615303,
|
| 636 |
"step": 5250
|
| 637 |
}
|
| 638 |
],
|
| 639 |
"logging_steps": 100,
|
| 640 |
"max_steps": 12500,
|
|
|
|
| 641 |
"num_train_epochs": 50,
|
| 642 |
"save_steps": 500,
|
| 643 |
+
"total_flos": 3.3566494603555635e+17,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 644 |
"trial_name": null,
|
| 645 |
"trial_params": null
|
| 646 |
}
|
checkpoint-5250/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 sha256:814f378bbfbf215f6c033d7b267886233699f8ce3ebf0a109279893fc2f542bf
|
| 3 |
+
size 4728
|