Training in progress, epoch 26, checkpoint
Browse files- checkpoint-6500/adapter_config.json +1 -1
- checkpoint-6500/adapter_model.bin +3 -0
- checkpoint-6500/added_tokens.json +3 -3
- checkpoint-6500/optimizer.pt +2 -2
- checkpoint-6500/rng_state_0.pth +1 -1
- checkpoint-6500/rng_state_1.pth +1 -1
- checkpoint-6500/rng_state_2.pth +1 -1
- checkpoint-6500/rng_state_3.pth +1 -1
- checkpoint-6500/special_tokens_map.json +4 -12
- checkpoint-6500/tokenizer.json +0 -0
- checkpoint-6500/tokenizer.model +3 -0
- checkpoint-6500/tokenizer_config.json +17 -20
- checkpoint-6500/trainer_state.json +248 -327
- checkpoint-6500/training_args.bin +2 -2
checkpoint-6500/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-6500/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ae69f0a0e8ee08e26d591e3627299e246f275ca385b8d889b999d850e33bf17d
|
| 3 |
+
size 143269386
|
checkpoint-6500/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-6500/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 sha256:38222fcf7ae92298629dc04fc0cfe421880f79f626593b83c39b4923a3524ade
|
| 3 |
+
size 286585234
|
checkpoint-6500/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 sha256:59ff9f8e4ed6601f079ebb3b627bee1855e188e2e2af68cb4fcdb8f99651efc4
|
| 3 |
size 15024
|
checkpoint-6500/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 sha256:668045c7c31f201281bd8a471b1d34ac4b5c327b3de0ff674e193450a58a52ce
|
| 3 |
size 15024
|
checkpoint-6500/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 sha256:861e2f2d6619581f53afc4f06ad03ac9bd103df66e1fc127ac77ac6f84ee194b
|
| 3 |
size 15024
|
checkpoint-6500/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 sha256:1d573959853d5498a1fcabee466b2cbf888e328adbaa90add7b62c002be274bb
|
| 3 |
size 15024
|
checkpoint-6500/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-6500/tokenizer.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-6500/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
checkpoint-6500/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-6500/trainer_state.json
CHANGED
|
@@ -10,869 +10,790 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.4,
|
| 13 |
-
"grad_norm": 0.48472875356674194,
|
| 14 |
"learning_rate": 0.0003,
|
| 15 |
-
"loss":
|
| 16 |
"step": 100
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.8,
|
| 20 |
-
"grad_norm": 0.4583953619003296,
|
| 21 |
"learning_rate": 0.0003,
|
| 22 |
-
"loss":
|
| 23 |
"step": 200
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 1.0,
|
| 27 |
-
"eval_accuracy": 0.
|
| 28 |
-
"eval_loss":
|
| 29 |
-
"eval_runtime": 5.
|
| 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": 0.6610094904899597,
|
| 43 |
"learning_rate": 0.0003,
|
| 44 |
-
"loss":
|
| 45 |
"step": 300
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 1.6,
|
| 49 |
-
"grad_norm": 0.7361629605293274,
|
| 50 |
"learning_rate": 0.0003,
|
| 51 |
-
"loss": 1.
|
| 52 |
"step": 400
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 2.0,
|
| 56 |
-
"grad_norm": 0.6793610453605652,
|
| 57 |
"learning_rate": 0.0003,
|
| 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": 0.8843489289283752,
|
| 79 |
"learning_rate": 0.0003,
|
| 80 |
-
"loss": 1.
|
| 81 |
"step": 600
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 2.8,
|
| 85 |
-
"grad_norm": 0.8380613923072815,
|
| 86 |
"learning_rate": 0.0003,
|
| 87 |
-
"loss": 1.
|
| 88 |
"step": 700
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 3.0,
|
| 92 |
-
"eval_accuracy": 0.
|
| 93 |
-
"eval_loss":
|
| 94 |
-
"eval_runtime": 5.
|
| 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": 1.068247675895691,
|
| 108 |
"learning_rate": 0.0003,
|
| 109 |
-
"loss": 1.
|
| 110 |
"step": 800
|
| 111 |
},
|
| 112 |
{
|
| 113 |
"epoch": 3.6,
|
| 114 |
-
"grad_norm": 1.041741967201233,
|
| 115 |
"learning_rate": 0.0003,
|
| 116 |
-
"loss":
|
| 117 |
"step": 900
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 4.0,
|
| 121 |
-
"grad_norm": 1.267968773841858,
|
| 122 |
"learning_rate": 0.0003,
|
| 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":
|
| 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": 1.1971005201339722,
|
| 144 |
"learning_rate": 0.0003,
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 1100
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 4.8,
|
| 150 |
-
"grad_norm": 1.2383127212524414,
|
| 151 |
"learning_rate": 0.0003,
|
| 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": 0.9558005928993225,
|
| 173 |
"learning_rate": 0.0003,
|
| 174 |
-
"loss": 0.
|
| 175 |
"step": 1300
|
| 176 |
},
|
| 177 |
{
|
| 178 |
"epoch": 5.6,
|
| 179 |
-
"grad_norm": 1.1341361999511719,
|
| 180 |
"learning_rate": 0.0003,
|
| 181 |
-
"loss": 0.
|
| 182 |
"step": 1400
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 6.0,
|
| 186 |
-
"grad_norm": 1.1375033855438232,
|
| 187 |
"learning_rate": 0.0003,
|
| 188 |
-
"loss": 0.
|
| 189 |
"step": 1500
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 6.0,
|
| 193 |
-
"eval_accuracy": 0.
|
| 194 |
-
"eval_loss":
|
| 195 |
-
"eval_runtime":
|
| 196 |
-
"eval_samples_per_second":
|
| 197 |
-
"eval_steps_per_second":
|
| 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": 1.118528962135315,
|
| 209 |
"learning_rate": 0.0003,
|
| 210 |
-
"loss": 0.
|
| 211 |
"step": 1600
|
| 212 |
},
|
| 213 |
{
|
| 214 |
"epoch": 6.8,
|
| 215 |
-
"grad_norm": 0.9311291575431824,
|
| 216 |
"learning_rate": 0.0003,
|
| 217 |
-
"loss": 0.
|
| 218 |
"step": 1700
|
| 219 |
},
|
| 220 |
{
|
| 221 |
"epoch": 7.0,
|
| 222 |
-
"eval_accuracy": 0.
|
| 223 |
-
"eval_loss":
|
| 224 |
-
"eval_runtime": 5.
|
| 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": 1.0299904346466064,
|
| 238 |
"learning_rate": 0.0003,
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 1800
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 7.6,
|
| 244 |
-
"grad_norm": 0.7518107295036316,
|
| 245 |
"learning_rate": 0.0003,
|
| 246 |
-
"loss": 0.
|
| 247 |
"step": 1900
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"epoch": 8.0,
|
| 251 |
-
"grad_norm": 1.1140515804290771,
|
| 252 |
"learning_rate": 0.0003,
|
| 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": 1.2363754510879517,
|
| 274 |
"learning_rate": 0.0003,
|
| 275 |
-
"loss": 0.
|
| 276 |
"step": 2100
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 8.8,
|
| 280 |
-
"grad_norm": 0.8174892067909241,
|
| 281 |
"learning_rate": 0.0003,
|
| 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": 0.551261842250824,
|
| 303 |
"learning_rate": 0.0003,
|
| 304 |
-
"loss": 0.
|
| 305 |
"step": 2300
|
| 306 |
},
|
| 307 |
{
|
| 308 |
"epoch": 9.6,
|
| 309 |
-
"grad_norm": 0.6361530423164368,
|
| 310 |
"learning_rate": 0.0003,
|
| 311 |
-
"loss": 0.
|
| 312 |
"step": 2400
|
| 313 |
},
|
| 314 |
{
|
| 315 |
"epoch": 10.0,
|
| 316 |
-
"grad_norm": 0.9359388947486877,
|
| 317 |
"learning_rate": 0.0003,
|
| 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": 0.7834428548812866,
|
| 339 |
"learning_rate": 0.0003,
|
| 340 |
-
"loss": 0.
|
| 341 |
"step": 2600
|
| 342 |
},
|
| 343 |
{
|
| 344 |
"epoch": 10.8,
|
| 345 |
-
"grad_norm": 1.0201836824417114,
|
| 346 |
"learning_rate": 0.0003,
|
| 347 |
-
"loss": 0.
|
| 348 |
"step": 2700
|
| 349 |
},
|
| 350 |
{
|
| 351 |
"epoch": 11.0,
|
| 352 |
-
"eval_accuracy": 0.
|
| 353 |
-
"eval_loss": 3.
|
| 354 |
-
"eval_runtime":
|
| 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": 0.738467812538147,
|
| 368 |
"learning_rate": 0.0003,
|
| 369 |
-
"loss": 0.
|
| 370 |
"step": 2800
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 11.6,
|
| 374 |
-
"grad_norm": 0.535826563835144,
|
| 375 |
"learning_rate": 0.0003,
|
| 376 |
-
"loss": 0.
|
| 377 |
"step": 2900
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 12.0,
|
| 381 |
-
"grad_norm": 0.8586340546607971,
|
| 382 |
"learning_rate": 0.0003,
|
| 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": 0.6024760007858276,
|
| 404 |
"learning_rate": 0.0003,
|
| 405 |
-
"loss": 0.
|
| 406 |
"step": 3100
|
| 407 |
},
|
| 408 |
{
|
| 409 |
"epoch": 12.8,
|
| 410 |
-
"grad_norm": 0.7764168381690979,
|
| 411 |
"learning_rate": 0.0003,
|
| 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": 0.668953537940979,
|
| 433 |
"learning_rate": 0.0003,
|
| 434 |
-
"loss": 0.
|
| 435 |
"step": 3300
|
| 436 |
},
|
| 437 |
{
|
| 438 |
"epoch": 13.6,
|
| 439 |
-
"grad_norm": 0.7968106865882874,
|
| 440 |
"learning_rate": 0.0003,
|
| 441 |
-
"loss": 0.
|
| 442 |
"step": 3400
|
| 443 |
},
|
| 444 |
{
|
| 445 |
"epoch": 14.0,
|
| 446 |
-
"grad_norm": 0.7598584294319153,
|
| 447 |
"learning_rate": 0.0003,
|
| 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": 0.7232977151870728,
|
| 469 |
"learning_rate": 0.0003,
|
| 470 |
-
"loss": 0.
|
| 471 |
"step": 3600
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 14.8,
|
| 475 |
-
"grad_norm": 0.6823480725288391,
|
| 476 |
"learning_rate": 0.0003,
|
| 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": 0.5730023384094238,
|
| 498 |
"learning_rate": 0.0003,
|
| 499 |
-
"loss": 0.
|
| 500 |
"step": 3800
|
| 501 |
},
|
| 502 |
{
|
| 503 |
"epoch": 15.6,
|
| 504 |
-
"grad_norm": 0.943156898021698,
|
| 505 |
"learning_rate": 0.0003,
|
| 506 |
-
"loss": 0.
|
| 507 |
"step": 3900
|
| 508 |
},
|
| 509 |
{
|
| 510 |
"epoch": 16.0,
|
| 511 |
-
"grad_norm": 0.8560875654220581,
|
| 512 |
"learning_rate": 0.0003,
|
| 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": 0.5993083715438843,
|
| 534 |
"learning_rate": 0.0003,
|
| 535 |
-
"loss": 0.
|
| 536 |
"step": 4100
|
| 537 |
},
|
| 538 |
{
|
| 539 |
"epoch": 16.8,
|
| 540 |
-
"grad_norm": 0.8744499683380127,
|
| 541 |
"learning_rate": 0.0003,
|
| 542 |
-
"loss": 0.
|
| 543 |
"step": 4200
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 17.0,
|
| 547 |
-
"eval_accuracy": 0.
|
| 548 |
-
"eval_loss": 3.
|
| 549 |
-
"eval_runtime":
|
| 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": 0.7331182360649109,
|
| 563 |
"learning_rate": 0.0003,
|
| 564 |
-
"loss": 0.
|
| 565 |
"step": 4300
|
| 566 |
},
|
| 567 |
{
|
| 568 |
"epoch": 17.6,
|
| 569 |
-
"grad_norm": 0.9493301510810852,
|
| 570 |
"learning_rate": 0.0003,
|
| 571 |
-
"loss": 0.
|
| 572 |
"step": 4400
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 18.0,
|
| 576 |
-
"grad_norm": 0.7808287143707275,
|
| 577 |
"learning_rate": 0.0003,
|
| 578 |
-
"loss": 0.
|
| 579 |
"step": 4500
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 18.0,
|
| 583 |
-
"eval_accuracy": 0.
|
| 584 |
-
"eval_loss": 3.
|
| 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": 0.7472721934318542,
|
| 599 |
"learning_rate": 0.0003,
|
| 600 |
-
"loss": 0.
|
| 601 |
"step": 4600
|
| 602 |
},
|
| 603 |
{
|
| 604 |
"epoch": 18.8,
|
| 605 |
-
"grad_norm": 0.5398291349411011,
|
| 606 |
"learning_rate": 0.0003,
|
| 607 |
-
"loss": 0.
|
| 608 |
"step": 4700
|
| 609 |
},
|
| 610 |
{
|
| 611 |
"epoch": 19.0,
|
| 612 |
-
"eval_accuracy": 0.
|
| 613 |
-
"eval_loss": 3.
|
| 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": 0.5717331171035767,
|
| 628 |
"learning_rate": 0.0003,
|
| 629 |
-
"loss": 0.
|
| 630 |
"step": 4800
|
| 631 |
},
|
| 632 |
{
|
| 633 |
"epoch": 19.6,
|
| 634 |
-
"grad_norm": 0.8405028581619263,
|
| 635 |
"learning_rate": 0.0003,
|
| 636 |
-
"loss": 0.
|
| 637 |
"step": 4900
|
| 638 |
},
|
| 639 |
{
|
| 640 |
"epoch": 20.0,
|
| 641 |
-
"grad_norm": 0.7853202223777771,
|
| 642 |
"learning_rate": 0.0003,
|
| 643 |
-
"loss": 0.
|
| 644 |
"step": 5000
|
| 645 |
},
|
| 646 |
{
|
| 647 |
"epoch": 20.0,
|
| 648 |
-
"eval_accuracy": 0.
|
| 649 |
-
"eval_loss":
|
| 650 |
-
"eval_runtime":
|
| 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": 0.8046431541442871,
|
| 664 |
"learning_rate": 0.0003,
|
| 665 |
-
"loss": 0.
|
| 666 |
"step": 5100
|
| 667 |
},
|
| 668 |
{
|
| 669 |
"epoch": 20.8,
|
| 670 |
-
"grad_norm": 0.6186045408248901,
|
| 671 |
"learning_rate": 0.0003,
|
| 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 |
"epoch": 21.2,
|
| 692 |
-
"grad_norm": 0.4458799660205841,
|
| 693 |
"learning_rate": 0.0003,
|
| 694 |
-
"loss": 0.
|
| 695 |
"step": 5300
|
| 696 |
},
|
| 697 |
{
|
| 698 |
"epoch": 21.6,
|
| 699 |
-
"grad_norm": 0.6755269765853882,
|
| 700 |
"learning_rate": 0.0003,
|
| 701 |
-
"loss": 0.
|
| 702 |
"step": 5400
|
| 703 |
},
|
| 704 |
{
|
| 705 |
"epoch": 22.0,
|
| 706 |
-
"grad_norm": 0.8081554174423218,
|
| 707 |
"learning_rate": 0.0003,
|
| 708 |
-
"loss": 0.
|
| 709 |
"step": 5500
|
| 710 |
},
|
| 711 |
{
|
| 712 |
"epoch": 22.0,
|
| 713 |
-
"eval_accuracy": 0.
|
| 714 |
-
"eval_loss": 3.
|
| 715 |
-
"eval_runtime":
|
| 716 |
-
"eval_samples_per_second":
|
| 717 |
-
"eval_steps_per_second":
|
| 718 |
"step": 5500
|
| 719 |
},
|
| 720 |
{
|
| 721 |
"epoch": 22.0,
|
| 722 |
-
"eval_exact_match":
|
| 723 |
-
"eval_f1":
|
| 724 |
"step": 5500
|
| 725 |
},
|
| 726 |
{
|
| 727 |
"epoch": 22.4,
|
| 728 |
-
"grad_norm": 0.5928174257278442,
|
| 729 |
"learning_rate": 0.0003,
|
| 730 |
-
"loss": 0.
|
| 731 |
"step": 5600
|
| 732 |
},
|
| 733 |
{
|
| 734 |
"epoch": 22.8,
|
| 735 |
-
"grad_norm": 0.5147334933280945,
|
| 736 |
"learning_rate": 0.0003,
|
| 737 |
-
"loss": 0.
|
| 738 |
"step": 5700
|
| 739 |
},
|
| 740 |
{
|
| 741 |
"epoch": 23.0,
|
| 742 |
-
"eval_accuracy": 0.
|
| 743 |
-
"eval_loss": 3.
|
| 744 |
-
"eval_runtime":
|
| 745 |
-
"eval_samples_per_second":
|
| 746 |
-
"eval_steps_per_second":
|
| 747 |
"step": 5750
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 23.0,
|
| 751 |
-
"eval_exact_match":
|
| 752 |
-
"eval_f1":
|
| 753 |
"step": 5750
|
| 754 |
},
|
| 755 |
{
|
| 756 |
"epoch": 23.2,
|
| 757 |
-
"grad_norm": 0.8577277660369873,
|
| 758 |
"learning_rate": 0.0003,
|
| 759 |
-
"loss": 0.
|
| 760 |
"step": 5800
|
| 761 |
},
|
| 762 |
{
|
| 763 |
"epoch": 23.6,
|
| 764 |
-
"grad_norm": 0.6065458655357361,
|
| 765 |
"learning_rate": 0.0003,
|
| 766 |
-
"loss": 0.
|
| 767 |
"step": 5900
|
| 768 |
},
|
| 769 |
{
|
| 770 |
"epoch": 24.0,
|
| 771 |
-
"grad_norm": 1.1300039291381836,
|
| 772 |
"learning_rate": 0.0003,
|
| 773 |
-
"loss": 0.
|
| 774 |
"step": 6000
|
| 775 |
},
|
| 776 |
{
|
| 777 |
"epoch": 24.0,
|
| 778 |
-
"eval_accuracy": 0.
|
| 779 |
-
"eval_loss": 3.
|
| 780 |
-
"eval_runtime":
|
| 781 |
-
"eval_samples_per_second":
|
| 782 |
-
"eval_steps_per_second":
|
| 783 |
"step": 6000
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 24.0,
|
| 787 |
-
"eval_exact_match":
|
| 788 |
-
"eval_f1":
|
| 789 |
"step": 6000
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 24.4,
|
| 793 |
-
"grad_norm": 0.722024142742157,
|
| 794 |
"learning_rate": 0.0003,
|
| 795 |
-
"loss": 0.
|
| 796 |
"step": 6100
|
| 797 |
},
|
| 798 |
{
|
| 799 |
"epoch": 24.8,
|
| 800 |
-
"grad_norm": 0.773781955242157,
|
| 801 |
"learning_rate": 0.0003,
|
| 802 |
-
"loss": 0.
|
| 803 |
"step": 6200
|
| 804 |
},
|
| 805 |
{
|
| 806 |
"epoch": 25.0,
|
| 807 |
-
"eval_accuracy": 0.
|
| 808 |
-
"eval_loss": 3.
|
| 809 |
-
"eval_runtime":
|
| 810 |
-
"eval_samples_per_second":
|
| 811 |
-
"eval_steps_per_second":
|
| 812 |
"step": 6250
|
| 813 |
},
|
| 814 |
{
|
| 815 |
"epoch": 25.0,
|
| 816 |
-
"eval_exact_match":
|
| 817 |
-
"eval_f1":
|
| 818 |
"step": 6250
|
| 819 |
},
|
| 820 |
{
|
| 821 |
"epoch": 25.2,
|
| 822 |
-
"grad_norm": 0.5301027894020081,
|
| 823 |
"learning_rate": 0.0003,
|
| 824 |
-
"loss": 0.
|
| 825 |
"step": 6300
|
| 826 |
},
|
| 827 |
{
|
| 828 |
"epoch": 25.6,
|
| 829 |
-
"grad_norm": 0.6550167202949524,
|
| 830 |
"learning_rate": 0.0003,
|
| 831 |
-
"loss": 0.
|
| 832 |
"step": 6400
|
| 833 |
},
|
| 834 |
{
|
| 835 |
"epoch": 26.0,
|
| 836 |
-
"grad_norm": 0.9597300291061401,
|
| 837 |
"learning_rate": 0.0003,
|
| 838 |
-
"loss": 0.
|
| 839 |
"step": 6500
|
| 840 |
},
|
| 841 |
{
|
| 842 |
"epoch": 26.0,
|
| 843 |
-
"eval_accuracy": 0.
|
| 844 |
-
"eval_loss": 3.
|
| 845 |
-
"eval_runtime":
|
| 846 |
-
"eval_samples_per_second":
|
| 847 |
-
"eval_steps_per_second":
|
| 848 |
"step": 6500
|
| 849 |
},
|
| 850 |
{
|
| 851 |
"epoch": 26.0,
|
| 852 |
-
"eval_exact_match":
|
| 853 |
-
"eval_f1":
|
| 854 |
"step": 6500
|
| 855 |
}
|
| 856 |
],
|
| 857 |
"logging_steps": 100,
|
| 858 |
"max_steps": 12500,
|
| 859 |
-
"num_input_tokens_seen": 0,
|
| 860 |
"num_train_epochs": 50,
|
| 861 |
"save_steps": 500,
|
| 862 |
-
"
|
| 863 |
-
"TrainerControl": {
|
| 864 |
-
"args": {
|
| 865 |
-
"should_epoch_stop": false,
|
| 866 |
-
"should_evaluate": false,
|
| 867 |
-
"should_log": false,
|
| 868 |
-
"should_save": true,
|
| 869 |
-
"should_training_stop": false
|
| 870 |
-
},
|
| 871 |
-
"attributes": {}
|
| 872 |
-
}
|
| 873 |
-
},
|
| 874 |
-
"total_flos": 1.6816001702166528e+17,
|
| 875 |
-
"train_batch_size": 1,
|
| 876 |
"trial_name": null,
|
| 877 |
"trial_params": null
|
| 878 |
}
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.4,
|
|
|
|
| 13 |
"learning_rate": 0.0003,
|
| 14 |
+
"loss": 1.9296,
|
| 15 |
"step": 100
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.8,
|
|
|
|
| 19 |
"learning_rate": 0.0003,
|
| 20 |
+
"loss": 1.7554,
|
| 21 |
"step": 200
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 1.0,
|
| 25 |
+
"eval_accuracy": 0.6093164556962025,
|
| 26 |
+
"eval_loss": 1.7939746379852295,
|
| 27 |
+
"eval_runtime": 5.0177,
|
| 28 |
+
"eval_samples_per_second": 99.648,
|
| 29 |
+
"eval_steps_per_second": 12.556,
|
| 30 |
"step": 250
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 1.0,
|
| 34 |
+
"eval_exact_match": 26.0,
|
| 35 |
+
"eval_f1": 36.71136363636365,
|
| 36 |
"step": 250
|
| 37 |
},
|
| 38 |
{
|
| 39 |
"epoch": 1.2,
|
|
|
|
| 40 |
"learning_rate": 0.0003,
|
| 41 |
+
"loss": 1.6394,
|
| 42 |
"step": 300
|
| 43 |
},
|
| 44 |
{
|
| 45 |
"epoch": 1.6,
|
|
|
|
| 46 |
"learning_rate": 0.0003,
|
| 47 |
+
"loss": 1.5315,
|
| 48 |
"step": 400
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 2.0,
|
|
|
|
| 52 |
"learning_rate": 0.0003,
|
| 53 |
+
"loss": 1.5248,
|
| 54 |
"step": 500
|
| 55 |
},
|
| 56 |
{
|
| 57 |
"epoch": 2.0,
|
| 58 |
+
"eval_accuracy": 0.6085063291139241,
|
| 59 |
+
"eval_loss": 1.8273799419403076,
|
| 60 |
+
"eval_runtime": 4.4023,
|
| 61 |
+
"eval_samples_per_second": 113.576,
|
| 62 |
+
"eval_steps_per_second": 14.311,
|
| 63 |
"step": 500
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 2.0,
|
| 67 |
+
"eval_exact_match": 24.4,
|
| 68 |
+
"eval_f1": 33.57723554223556,
|
| 69 |
"step": 500
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.4,
|
|
|
|
| 73 |
"learning_rate": 0.0003,
|
| 74 |
+
"loss": 1.1692,
|
| 75 |
"step": 600
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 2.8,
|
|
|
|
| 79 |
"learning_rate": 0.0003,
|
| 80 |
+
"loss": 1.2054,
|
| 81 |
"step": 700
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 3.0,
|
| 85 |
+
"eval_accuracy": 0.6027088607594937,
|
| 86 |
+
"eval_loss": 1.9717934131622314,
|
| 87 |
+
"eval_runtime": 5.1273,
|
| 88 |
+
"eval_samples_per_second": 97.518,
|
| 89 |
+
"eval_steps_per_second": 12.287,
|
| 90 |
"step": 750
|
| 91 |
},
|
| 92 |
{
|
| 93 |
"epoch": 3.0,
|
| 94 |
+
"eval_exact_match": 20.8,
|
| 95 |
+
"eval_f1": 30.584243697478996,
|
| 96 |
"step": 750
|
| 97 |
},
|
| 98 |
{
|
| 99 |
"epoch": 3.2,
|
|
|
|
| 100 |
"learning_rate": 0.0003,
|
| 101 |
+
"loss": 1.0252,
|
| 102 |
"step": 800
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 3.6,
|
|
|
|
| 106 |
"learning_rate": 0.0003,
|
| 107 |
+
"loss": 0.8608,
|
| 108 |
"step": 900
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 4.0,
|
|
|
|
| 112 |
"learning_rate": 0.0003,
|
| 113 |
+
"loss": 0.8989,
|
| 114 |
"step": 1000
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 4.0,
|
| 118 |
+
"eval_accuracy": 0.5986835443037974,
|
| 119 |
+
"eval_loss": 2.151914596557617,
|
| 120 |
+
"eval_runtime": 5.3325,
|
| 121 |
+
"eval_samples_per_second": 93.765,
|
| 122 |
+
"eval_steps_per_second": 11.814,
|
| 123 |
"step": 1000
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 4.0,
|
| 127 |
+
"eval_exact_match": 21.0,
|
| 128 |
+
"eval_f1": 30.363229548229558,
|
| 129 |
"step": 1000
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 4.4,
|
|
|
|
| 133 |
"learning_rate": 0.0003,
|
| 134 |
+
"loss": 0.5842,
|
| 135 |
"step": 1100
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 4.8,
|
|
|
|
| 139 |
"learning_rate": 0.0003,
|
| 140 |
+
"loss": 0.6306,
|
| 141 |
"step": 1200
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"epoch": 5.0,
|
| 145 |
+
"eval_accuracy": 0.5960506329113924,
|
| 146 |
+
"eval_loss": 2.329284191131592,
|
| 147 |
+
"eval_runtime": 4.5585,
|
| 148 |
+
"eval_samples_per_second": 109.686,
|
| 149 |
+
"eval_steps_per_second": 13.82,
|
| 150 |
"step": 1250
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 5.0,
|
| 154 |
+
"eval_exact_match": 20.4,
|
| 155 |
+
"eval_f1": 30.02542624042625,
|
| 156 |
"step": 1250
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 5.2,
|
|
|
|
| 160 |
"learning_rate": 0.0003,
|
| 161 |
+
"loss": 0.5368,
|
| 162 |
"step": 1300
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 5.6,
|
|
|
|
| 166 |
"learning_rate": 0.0003,
|
| 167 |
+
"loss": 0.4495,
|
| 168 |
"step": 1400
|
| 169 |
},
|
| 170 |
{
|
| 171 |
"epoch": 6.0,
|
|
|
|
| 172 |
"learning_rate": 0.0003,
|
| 173 |
+
"loss": 0.4712,
|
| 174 |
"step": 1500
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 6.0,
|
| 178 |
+
"eval_accuracy": 0.5935696202531645,
|
| 179 |
+
"eval_loss": 2.5598793029785156,
|
| 180 |
+
"eval_runtime": 4.4094,
|
| 181 |
+
"eval_samples_per_second": 113.395,
|
| 182 |
+
"eval_steps_per_second": 14.288,
|
| 183 |
"step": 1500
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 6.0,
|
| 187 |
+
"eval_exact_match": 20.6,
|
| 188 |
+
"eval_f1": 30.108149073149082,
|
| 189 |
"step": 1500
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 6.4,
|
|
|
|
| 193 |
"learning_rate": 0.0003,
|
| 194 |
+
"loss": 0.3625,
|
| 195 |
"step": 1600
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 6.8,
|
|
|
|
| 199 |
"learning_rate": 0.0003,
|
| 200 |
+
"loss": 0.3797,
|
| 201 |
"step": 1700
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 7.0,
|
| 205 |
+
"eval_accuracy": 0.5935696202531645,
|
| 206 |
+
"eval_loss": 2.732860803604126,
|
| 207 |
+
"eval_runtime": 5.2701,
|
| 208 |
+
"eval_samples_per_second": 94.876,
|
| 209 |
+
"eval_steps_per_second": 11.954,
|
| 210 |
"step": 1750
|
| 211 |
},
|
| 212 |
{
|
| 213 |
"epoch": 7.0,
|
| 214 |
+
"eval_exact_match": 19.6,
|
| 215 |
+
"eval_f1": 29.002709512709526,
|
| 216 |
"step": 1750
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 7.2,
|
|
|
|
| 220 |
"learning_rate": 0.0003,
|
| 221 |
+
"loss": 0.3582,
|
| 222 |
"step": 1800
|
| 223 |
},
|
| 224 |
{
|
| 225 |
"epoch": 7.6,
|
|
|
|
| 226 |
"learning_rate": 0.0003,
|
| 227 |
+
"loss": 0.3399,
|
| 228 |
"step": 1900
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 8.0,
|
|
|
|
| 232 |
"learning_rate": 0.0003,
|
| 233 |
+
"loss": 0.3527,
|
| 234 |
"step": 2000
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 8.0,
|
| 238 |
+
"eval_accuracy": 0.5912911392405064,
|
| 239 |
+
"eval_loss": 2.8185083866119385,
|
| 240 |
+
"eval_runtime": 4.4105,
|
| 241 |
+
"eval_samples_per_second": 113.366,
|
| 242 |
+
"eval_steps_per_second": 14.284,
|
| 243 |
"step": 2000
|
| 244 |
},
|
| 245 |
{
|
| 246 |
"epoch": 8.0,
|
| 247 |
+
"eval_exact_match": 20.8,
|
| 248 |
+
"eval_f1": 29.191525696525694,
|
| 249 |
"step": 2000
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 8.4,
|
|
|
|
| 253 |
"learning_rate": 0.0003,
|
| 254 |
+
"loss": 0.3202,
|
| 255 |
"step": 2100
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 8.8,
|
|
|
|
| 259 |
"learning_rate": 0.0003,
|
| 260 |
+
"loss": 0.3314,
|
| 261 |
"step": 2200
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 9.0,
|
| 265 |
+
"eval_accuracy": 0.592,
|
| 266 |
+
"eval_loss": 2.824962854385376,
|
| 267 |
+
"eval_runtime": 4.5596,
|
| 268 |
+
"eval_samples_per_second": 109.66,
|
| 269 |
+
"eval_steps_per_second": 13.817,
|
| 270 |
"step": 2250
|
| 271 |
},
|
| 272 |
{
|
| 273 |
"epoch": 9.0,
|
| 274 |
+
"eval_exact_match": 21.0,
|
| 275 |
+
"eval_f1": 29.434571539571547,
|
| 276 |
"step": 2250
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 9.2,
|
|
|
|
| 280 |
"learning_rate": 0.0003,
|
| 281 |
+
"loss": 0.3174,
|
| 282 |
"step": 2300
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 9.6,
|
|
|
|
| 286 |
"learning_rate": 0.0003,
|
| 287 |
+
"loss": 0.3157,
|
| 288 |
"step": 2400
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 10.0,
|
|
|
|
| 292 |
"learning_rate": 0.0003,
|
| 293 |
+
"loss": 0.3265,
|
| 294 |
"step": 2500
|
| 295 |
},
|
| 296 |
{
|
| 297 |
"epoch": 10.0,
|
| 298 |
+
"eval_accuracy": 0.5911392405063292,
|
| 299 |
+
"eval_loss": 2.9242382049560547,
|
| 300 |
+
"eval_runtime": 4.7857,
|
| 301 |
+
"eval_samples_per_second": 104.477,
|
| 302 |
+
"eval_steps_per_second": 13.164,
|
| 303 |
"step": 2500
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 10.0,
|
| 307 |
+
"eval_exact_match": 20.8,
|
| 308 |
+
"eval_f1": 29.524657009657016,
|
| 309 |
"step": 2500
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 10.4,
|
|
|
|
| 313 |
"learning_rate": 0.0003,
|
| 314 |
+
"loss": 0.2989,
|
| 315 |
"step": 2600
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"epoch": 10.8,
|
|
|
|
| 319 |
"learning_rate": 0.0003,
|
| 320 |
+
"loss": 0.3148,
|
| 321 |
"step": 2700
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 11.0,
|
| 325 |
+
"eval_accuracy": 0.591240506329114,
|
| 326 |
+
"eval_loss": 3.0012593269348145,
|
| 327 |
+
"eval_runtime": 4.5582,
|
| 328 |
+
"eval_samples_per_second": 109.691,
|
| 329 |
+
"eval_steps_per_second": 13.821,
|
| 330 |
"step": 2750
|
| 331 |
},
|
| 332 |
{
|
| 333 |
"epoch": 11.0,
|
| 334 |
+
"eval_exact_match": 20.0,
|
| 335 |
+
"eval_f1": 28.128880727116037,
|
| 336 |
"step": 2750
|
| 337 |
},
|
| 338 |
{
|
| 339 |
"epoch": 11.2,
|
|
|
|
| 340 |
"learning_rate": 0.0003,
|
| 341 |
+
"loss": 0.3028,
|
| 342 |
"step": 2800
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 11.6,
|
|
|
|
| 346 |
"learning_rate": 0.0003,
|
| 347 |
+
"loss": 0.3047,
|
| 348 |
"step": 2900
|
| 349 |
},
|
| 350 |
{
|
| 351 |
"epoch": 12.0,
|
|
|
|
| 352 |
"learning_rate": 0.0003,
|
| 353 |
+
"loss": 0.3184,
|
| 354 |
"step": 3000
|
| 355 |
},
|
| 356 |
{
|
| 357 |
"epoch": 12.0,
|
| 358 |
+
"eval_accuracy": 0.590632911392405,
|
| 359 |
+
"eval_loss": 2.931525707244873,
|
| 360 |
+
"eval_runtime": 4.4989,
|
| 361 |
+
"eval_samples_per_second": 111.138,
|
| 362 |
+
"eval_steps_per_second": 14.003,
|
| 363 |
"step": 3000
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 12.0,
|
| 367 |
+
"eval_exact_match": 21.2,
|
| 368 |
+
"eval_f1": 30.12248362748364,
|
| 369 |
"step": 3000
|
| 370 |
},
|
| 371 |
{
|
| 372 |
"epoch": 12.4,
|
|
|
|
| 373 |
"learning_rate": 0.0003,
|
| 374 |
+
"loss": 0.3,
|
| 375 |
"step": 3100
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 12.8,
|
|
|
|
| 379 |
"learning_rate": 0.0003,
|
| 380 |
+
"loss": 0.3101,
|
| 381 |
"step": 3200
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 13.0,
|
| 385 |
+
"eval_accuracy": 0.5896962025316456,
|
| 386 |
+
"eval_loss": 2.9116382598876953,
|
| 387 |
+
"eval_runtime": 4.6155,
|
| 388 |
+
"eval_samples_per_second": 108.331,
|
| 389 |
+
"eval_steps_per_second": 13.65,
|
| 390 |
"step": 3250
|
| 391 |
},
|
| 392 |
{
|
| 393 |
"epoch": 13.0,
|
| 394 |
+
"eval_exact_match": 19.2,
|
| 395 |
+
"eval_f1": 27.90991785991786,
|
| 396 |
"step": 3250
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 13.2,
|
|
|
|
| 400 |
"learning_rate": 0.0003,
|
| 401 |
+
"loss": 0.3063,
|
| 402 |
"step": 3300
|
| 403 |
},
|
| 404 |
{
|
| 405 |
"epoch": 13.6,
|
|
|
|
| 406 |
"learning_rate": 0.0003,
|
| 407 |
+
"loss": 0.3041,
|
| 408 |
"step": 3400
|
| 409 |
},
|
| 410 |
{
|
| 411 |
"epoch": 14.0,
|
|
|
|
| 412 |
"learning_rate": 0.0003,
|
| 413 |
+
"loss": 0.3164,
|
| 414 |
"step": 3500
|
| 415 |
},
|
| 416 |
{
|
| 417 |
"epoch": 14.0,
|
| 418 |
+
"eval_accuracy": 0.5902025316455696,
|
| 419 |
+
"eval_loss": 2.920793056488037,
|
| 420 |
+
"eval_runtime": 4.4066,
|
| 421 |
+
"eval_samples_per_second": 113.467,
|
| 422 |
+
"eval_steps_per_second": 14.297,
|
| 423 |
"step": 3500
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"epoch": 14.0,
|
| 427 |
+
"eval_exact_match": 18.0,
|
| 428 |
+
"eval_f1": 27.415170385170395,
|
| 429 |
"step": 3500
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 14.4,
|
|
|
|
| 433 |
"learning_rate": 0.0003,
|
| 434 |
+
"loss": 0.2957,
|
| 435 |
"step": 3600
|
| 436 |
},
|
| 437 |
{
|
| 438 |
"epoch": 14.8,
|
|
|
|
| 439 |
"learning_rate": 0.0003,
|
| 440 |
+
"loss": 0.3074,
|
| 441 |
"step": 3700
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 15.0,
|
| 445 |
+
"eval_accuracy": 0.5908607594936709,
|
| 446 |
+
"eval_loss": 2.9385440349578857,
|
| 447 |
+
"eval_runtime": 4.8175,
|
| 448 |
+
"eval_samples_per_second": 103.788,
|
| 449 |
+
"eval_steps_per_second": 13.077,
|
| 450 |
"step": 3750
|
| 451 |
},
|
| 452 |
{
|
| 453 |
"epoch": 15.0,
|
| 454 |
+
"eval_exact_match": 21.2,
|
| 455 |
+
"eval_f1": 29.687630702630706,
|
| 456 |
"step": 3750
|
| 457 |
},
|
| 458 |
{
|
| 459 |
"epoch": 15.2,
|
|
|
|
| 460 |
"learning_rate": 0.0003,
|
| 461 |
+
"loss": 0.3013,
|
| 462 |
"step": 3800
|
| 463 |
},
|
| 464 |
{
|
| 465 |
"epoch": 15.6,
|
|
|
|
| 466 |
"learning_rate": 0.0003,
|
| 467 |
+
"loss": 0.3002,
|
| 468 |
"step": 3900
|
| 469 |
},
|
| 470 |
{
|
| 471 |
"epoch": 16.0,
|
|
|
|
| 472 |
"learning_rate": 0.0003,
|
| 473 |
+
"loss": 0.3107,
|
| 474 |
"step": 4000
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 16.0,
|
| 478 |
+
"eval_accuracy": 0.5891645569620253,
|
| 479 |
+
"eval_loss": 2.9519243240356445,
|
| 480 |
+
"eval_runtime": 4.5619,
|
| 481 |
+
"eval_samples_per_second": 109.603,
|
| 482 |
+
"eval_steps_per_second": 13.81,
|
| 483 |
"step": 4000
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 16.0,
|
| 487 |
+
"eval_exact_match": 19.8,
|
| 488 |
+
"eval_f1": 28.670117105117104,
|
| 489 |
"step": 4000
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 16.4,
|
|
|
|
| 493 |
"learning_rate": 0.0003,
|
| 494 |
+
"loss": 0.2892,
|
| 495 |
"step": 4100
|
| 496 |
},
|
| 497 |
{
|
| 498 |
"epoch": 16.8,
|
|
|
|
| 499 |
"learning_rate": 0.0003,
|
| 500 |
+
"loss": 0.3054,
|
| 501 |
"step": 4200
|
| 502 |
},
|
| 503 |
{
|
| 504 |
"epoch": 17.0,
|
| 505 |
+
"eval_accuracy": 0.5898227848101266,
|
| 506 |
+
"eval_loss": 3.010847568511963,
|
| 507 |
+
"eval_runtime": 4.718,
|
| 508 |
+
"eval_samples_per_second": 105.977,
|
| 509 |
+
"eval_steps_per_second": 13.353,
|
| 510 |
"step": 4250
|
| 511 |
},
|
| 512 |
{
|
| 513 |
"epoch": 17.0,
|
| 514 |
+
"eval_exact_match": 19.6,
|
| 515 |
+
"eval_f1": 27.744197632432932,
|
| 516 |
"step": 4250
|
| 517 |
},
|
| 518 |
{
|
| 519 |
"epoch": 17.2,
|
|
|
|
| 520 |
"learning_rate": 0.0003,
|
| 521 |
+
"loss": 0.2959,
|
| 522 |
"step": 4300
|
| 523 |
},
|
| 524 |
{
|
| 525 |
"epoch": 17.6,
|
|
|
|
| 526 |
"learning_rate": 0.0003,
|
| 527 |
+
"loss": 0.297,
|
| 528 |
"step": 4400
|
| 529 |
},
|
| 530 |
{
|
| 531 |
"epoch": 18.0,
|
|
|
|
| 532 |
"learning_rate": 0.0003,
|
| 533 |
+
"loss": 0.309,
|
| 534 |
"step": 4500
|
| 535 |
},
|
| 536 |
{
|
| 537 |
"epoch": 18.0,
|
| 538 |
+
"eval_accuracy": 0.5903544303797469,
|
| 539 |
+
"eval_loss": 3.003683567047119,
|
| 540 |
+
"eval_runtime": 4.4085,
|
| 541 |
+
"eval_samples_per_second": 113.418,
|
| 542 |
+
"eval_steps_per_second": 14.291,
|
| 543 |
"step": 4500
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 18.0,
|
| 547 |
+
"eval_exact_match": 20.6,
|
| 548 |
+
"eval_f1": 28.64154694978225,
|
| 549 |
"step": 4500
|
| 550 |
},
|
| 551 |
{
|
| 552 |
"epoch": 18.4,
|
|
|
|
| 553 |
"learning_rate": 0.0003,
|
| 554 |
+
"loss": 0.2883,
|
| 555 |
"step": 4600
|
| 556 |
},
|
| 557 |
{
|
| 558 |
"epoch": 18.8,
|
|
|
|
| 559 |
"learning_rate": 0.0003,
|
| 560 |
+
"loss": 0.3005,
|
| 561 |
"step": 4700
|
| 562 |
},
|
| 563 |
{
|
| 564 |
"epoch": 19.0,
|
| 565 |
+
"eval_accuracy": 0.5898481012658228,
|
| 566 |
+
"eval_loss": 3.0279438495635986,
|
| 567 |
+
"eval_runtime": 4.7099,
|
| 568 |
+
"eval_samples_per_second": 106.16,
|
| 569 |
+
"eval_steps_per_second": 13.376,
|
| 570 |
"step": 4750
|
| 571 |
},
|
| 572 |
{
|
| 573 |
"epoch": 19.0,
|
| 574 |
+
"eval_exact_match": 21.0,
|
| 575 |
+
"eval_f1": 28.707790706026003,
|
| 576 |
"step": 4750
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 19.2,
|
|
|
|
| 580 |
"learning_rate": 0.0003,
|
| 581 |
+
"loss": 0.2959,
|
| 582 |
"step": 4800
|
| 583 |
},
|
| 584 |
{
|
| 585 |
"epoch": 19.6,
|
|
|
|
| 586 |
"learning_rate": 0.0003,
|
| 587 |
+
"loss": 0.2911,
|
| 588 |
"step": 4900
|
| 589 |
},
|
| 590 |
{
|
| 591 |
"epoch": 20.0,
|
|
|
|
| 592 |
"learning_rate": 0.0003,
|
| 593 |
+
"loss": 0.3127,
|
| 594 |
"step": 5000
|
| 595 |
},
|
| 596 |
{
|
| 597 |
"epoch": 20.0,
|
| 598 |
+
"eval_accuracy": 0.5883291139240506,
|
| 599 |
+
"eval_loss": 2.9650285243988037,
|
| 600 |
+
"eval_runtime": 4.711,
|
| 601 |
+
"eval_samples_per_second": 106.135,
|
| 602 |
+
"eval_steps_per_second": 13.373,
|
| 603 |
"step": 5000
|
| 604 |
},
|
| 605 |
{
|
| 606 |
"epoch": 20.0,
|
| 607 |
+
"eval_exact_match": 11.4,
|
| 608 |
+
"eval_f1": 20.32415197623728,
|
| 609 |
"step": 5000
|
| 610 |
},
|
| 611 |
{
|
| 612 |
"epoch": 20.4,
|
|
|
|
| 613 |
"learning_rate": 0.0003,
|
| 614 |
+
"loss": 0.2868,
|
| 615 |
"step": 5100
|
| 616 |
},
|
| 617 |
{
|
| 618 |
"epoch": 20.8,
|
|
|
|
| 619 |
"learning_rate": 0.0003,
|
| 620 |
+
"loss": 0.2989,
|
| 621 |
"step": 5200
|
| 622 |
},
|
| 623 |
{
|
| 624 |
"epoch": 21.0,
|
| 625 |
+
"eval_accuracy": 0.5895696202531645,
|
| 626 |
+
"eval_loss": 3.013432502746582,
|
| 627 |
+
"eval_runtime": 4.8101,
|
| 628 |
+
"eval_samples_per_second": 103.947,
|
| 629 |
+
"eval_steps_per_second": 13.097,
|
| 630 |
"step": 5250
|
| 631 |
},
|
| 632 |
{
|
| 633 |
"epoch": 21.0,
|
| 634 |
+
"eval_exact_match": 10.8,
|
| 635 |
+
"eval_f1": 17.55567536105352,
|
| 636 |
"step": 5250
|
| 637 |
},
|
| 638 |
{
|
| 639 |
"epoch": 21.2,
|
|
|
|
| 640 |
"learning_rate": 0.0003,
|
| 641 |
+
"loss": 0.2948,
|
| 642 |
"step": 5300
|
| 643 |
},
|
| 644 |
{
|
| 645 |
"epoch": 21.6,
|
|
|
|
| 646 |
"learning_rate": 0.0003,
|
| 647 |
+
"loss": 0.2921,
|
| 648 |
"step": 5400
|
| 649 |
},
|
| 650 |
{
|
| 651 |
"epoch": 22.0,
|
|
|
|
| 652 |
"learning_rate": 0.0003,
|
| 653 |
+
"loss": 0.3072,
|
| 654 |
"step": 5500
|
| 655 |
},
|
| 656 |
{
|
| 657 |
"epoch": 22.0,
|
| 658 |
+
"eval_accuracy": 0.5885822784810126,
|
| 659 |
+
"eval_loss": 3.0298521518707275,
|
| 660 |
+
"eval_runtime": 4.7113,
|
| 661 |
+
"eval_samples_per_second": 106.128,
|
| 662 |
+
"eval_steps_per_second": 13.372,
|
| 663 |
"step": 5500
|
| 664 |
},
|
| 665 |
{
|
| 666 |
"epoch": 22.0,
|
| 667 |
+
"eval_exact_match": 0.0,
|
| 668 |
+
"eval_f1": 0.7666666666666667,
|
| 669 |
"step": 5500
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 22.4,
|
|
|
|
| 673 |
"learning_rate": 0.0003,
|
| 674 |
+
"loss": 0.2914,
|
| 675 |
"step": 5600
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 22.8,
|
|
|
|
| 679 |
"learning_rate": 0.0003,
|
| 680 |
+
"loss": 0.2982,
|
| 681 |
"step": 5700
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 23.0,
|
| 685 |
+
"eval_accuracy": 0.5870126582278481,
|
| 686 |
+
"eval_loss": 3.0277092456817627,
|
| 687 |
+
"eval_runtime": 4.4065,
|
| 688 |
+
"eval_samples_per_second": 113.468,
|
| 689 |
+
"eval_steps_per_second": 14.297,
|
| 690 |
"step": 5750
|
| 691 |
},
|
| 692 |
{
|
| 693 |
"epoch": 23.0,
|
| 694 |
+
"eval_exact_match": 7.6,
|
| 695 |
+
"eval_f1": 9.586507936507935,
|
| 696 |
"step": 5750
|
| 697 |
},
|
| 698 |
{
|
| 699 |
"epoch": 23.2,
|
|
|
|
| 700 |
"learning_rate": 0.0003,
|
| 701 |
+
"loss": 0.3014,
|
| 702 |
"step": 5800
|
| 703 |
},
|
| 704 |
{
|
| 705 |
"epoch": 23.6,
|
|
|
|
| 706 |
"learning_rate": 0.0003,
|
| 707 |
+
"loss": 0.2907,
|
| 708 |
"step": 5900
|
| 709 |
},
|
| 710 |
{
|
| 711 |
"epoch": 24.0,
|
|
|
|
| 712 |
"learning_rate": 0.0003,
|
| 713 |
+
"loss": 0.3029,
|
| 714 |
"step": 6000
|
| 715 |
},
|
| 716 |
{
|
| 717 |
"epoch": 24.0,
|
| 718 |
+
"eval_accuracy": 0.5877215189873418,
|
| 719 |
+
"eval_loss": 3.02402925491333,
|
| 720 |
+
"eval_runtime": 4.7089,
|
| 721 |
+
"eval_samples_per_second": 106.182,
|
| 722 |
+
"eval_steps_per_second": 13.379,
|
| 723 |
"step": 6000
|
| 724 |
},
|
| 725 |
{
|
| 726 |
"epoch": 24.0,
|
| 727 |
+
"eval_exact_match": 4.2,
|
| 728 |
+
"eval_f1": 7.583745822888591,
|
| 729 |
"step": 6000
|
| 730 |
},
|
| 731 |
{
|
| 732 |
"epoch": 24.4,
|
|
|
|
| 733 |
"learning_rate": 0.0003,
|
| 734 |
+
"loss": 0.2853,
|
| 735 |
"step": 6100
|
| 736 |
},
|
| 737 |
{
|
| 738 |
"epoch": 24.8,
|
|
|
|
| 739 |
"learning_rate": 0.0003,
|
| 740 |
+
"loss": 0.2989,
|
| 741 |
"step": 6200
|
| 742 |
},
|
| 743 |
{
|
| 744 |
"epoch": 25.0,
|
| 745 |
+
"eval_accuracy": 0.5874430379746836,
|
| 746 |
+
"eval_loss": 3.0006086826324463,
|
| 747 |
+
"eval_runtime": 4.5617,
|
| 748 |
+
"eval_samples_per_second": 109.609,
|
| 749 |
+
"eval_steps_per_second": 13.811,
|
| 750 |
"step": 6250
|
| 751 |
},
|
| 752 |
{
|
| 753 |
"epoch": 25.0,
|
| 754 |
+
"eval_exact_match": 0.0,
|
| 755 |
+
"eval_f1": 0.5121500721500722,
|
| 756 |
"step": 6250
|
| 757 |
},
|
| 758 |
{
|
| 759 |
"epoch": 25.2,
|
|
|
|
| 760 |
"learning_rate": 0.0003,
|
| 761 |
+
"loss": 0.2904,
|
| 762 |
"step": 6300
|
| 763 |
},
|
| 764 |
{
|
| 765 |
"epoch": 25.6,
|
|
|
|
| 766 |
"learning_rate": 0.0003,
|
| 767 |
+
"loss": 0.29,
|
| 768 |
"step": 6400
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 26.0,
|
|
|
|
| 772 |
"learning_rate": 0.0003,
|
| 773 |
+
"loss": 0.3009,
|
| 774 |
"step": 6500
|
| 775 |
},
|
| 776 |
{
|
| 777 |
"epoch": 26.0,
|
| 778 |
+
"eval_accuracy": 0.5873417721518988,
|
| 779 |
+
"eval_loss": 3.058345317840576,
|
| 780 |
+
"eval_runtime": 4.708,
|
| 781 |
+
"eval_samples_per_second": 106.202,
|
| 782 |
+
"eval_steps_per_second": 13.381,
|
| 783 |
"step": 6500
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 26.0,
|
| 787 |
+
"eval_exact_match": 2.2,
|
| 788 |
+
"eval_f1": 4.8883016983016985,
|
| 789 |
"step": 6500
|
| 790 |
}
|
| 791 |
],
|
| 792 |
"logging_steps": 100,
|
| 793 |
"max_steps": 12500,
|
|
|
|
| 794 |
"num_train_epochs": 50,
|
| 795 |
"save_steps": 500,
|
| 796 |
+
"total_flos": 4.155661298769592e+17,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 797 |
"trial_name": null,
|
| 798 |
"trial_params": null
|
| 799 |
}
|
checkpoint-6500/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:eef995e0a4a8da1af8ae275504624b1fcb2d18639b8c53b553ebed1c9a9b7fd7
|
| 3 |
+
size 4728
|