Training in progress, epoch 21, checkpoint
Browse files- checkpoint-11109/adapter_config.json +1 -1
- checkpoint-11109/adapter_model.bin +3 -0
- checkpoint-11109/added_tokens.json +3 -3
- checkpoint-11109/optimizer.pt +2 -2
- checkpoint-11109/rng_state_0.pth +1 -1
- checkpoint-11109/rng_state_1.pth +1 -1
- checkpoint-11109/rng_state_2.pth +1 -1
- checkpoint-11109/rng_state_3.pth +1 -1
- checkpoint-11109/special_tokens_map.json +4 -12
- checkpoint-11109/tokenizer.json +0 -0
- checkpoint-11109/tokenizer.model +3 -0
- checkpoint-11109/tokenizer_config.json +17 -20
- checkpoint-11109/trainer_state.json +340 -465
- checkpoint-11109/training_args.bin +2 -2
checkpoint-11109/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-11109/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e85ade6c7ffd589a6d69cdef902199fa87c0d5c5fb9d367218e3af0711ab72b9
|
| 3 |
+
size 143269386
|
checkpoint-11109/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-11109/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 sha256:3aa557ef92812d47917a6154a16cc4177d20676a24c41d8ea36628e4694bfb82
|
| 3 |
+
size 286585234
|
checkpoint-11109/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 sha256:6952d4f54abbcd0959ab2aaa580f20ad0e492aba07145245de19261d90350794
|
| 3 |
size 15024
|
checkpoint-11109/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:eba86ae822bc6ca3fc40725bac6d1037fad3a0754e7d4a16c483095b629cb926
|
| 3 |
size 15024
|
checkpoint-11109/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 sha256:8334703c49aa82493b67357225d32a18b7fd1bf7dd150b9d52be94d3268fdc04
|
| 3 |
size 15024
|
checkpoint-11109/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:e8b356b6ec1ff810fd61d9593e5c1861362336d885cfba69e178a54738a18e1a
|
| 3 |
size 15024
|
checkpoint-11109/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-11109/tokenizer.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-11109/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
checkpoint-11109/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-11109/trainer_state.json
CHANGED
|
@@ -9,155 +9,140 @@
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm": 7.269421577453613,
|
| 14 |
"learning_rate": 0.0005,
|
| 15 |
-
"loss": 1.
|
| 16 |
"step": 100
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm": 1.071650505065918,
|
| 21 |
"learning_rate": 0.0005,
|
| 22 |
-
"loss": 1.
|
| 23 |
"step": 200
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm": 0.5662499070167542,
|
| 28 |
"learning_rate": 0.0005,
|
| 29 |
-
"loss":
|
| 30 |
"step": 300
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm": 0.47088223695755005,
|
| 35 |
"learning_rate": 0.0005,
|
| 36 |
-
"loss": 1.
|
| 37 |
"step": 400
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm": 0.3653932511806488,
|
| 42 |
"learning_rate": 0.0005,
|
| 43 |
-
"loss": 1.
|
| 44 |
"step": 500
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 1.0,
|
| 48 |
-
"eval_accuracy": 0.
|
| 49 |
-
"eval_loss": 1.
|
| 50 |
-
"eval_runtime":
|
| 51 |
-
"eval_samples_per_second":
|
| 52 |
-
"eval_steps_per_second":
|
| 53 |
"step": 529
|
| 54 |
},
|
| 55 |
{
|
| 56 |
"epoch": 1.0,
|
| 57 |
-
"eval_exact_match":
|
| 58 |
-
"eval_f1":
|
| 59 |
-
"eval_qa_bleu":
|
| 60 |
-
"eval_qa_exact_match": 0.
|
| 61 |
-
"eval_recite_bleu":
|
| 62 |
-
"eval_recite_exact_match": 0.
|
| 63 |
"step": 529
|
| 64 |
},
|
| 65 |
{
|
| 66 |
-
"epoch": 1.
|
| 67 |
-
"grad_norm": 0.5318541526794434,
|
| 68 |
"learning_rate": 0.0005,
|
| 69 |
-
"loss": 1.
|
| 70 |
"step": 600
|
| 71 |
},
|
| 72 |
{
|
| 73 |
-
"epoch": 1.
|
| 74 |
-
"grad_norm": 235.6529998779297,
|
| 75 |
"learning_rate": 0.0005,
|
| 76 |
-
"loss":
|
| 77 |
"step": 700
|
| 78 |
},
|
| 79 |
{
|
| 80 |
-
"epoch": 1.
|
| 81 |
-
"grad_norm": 0.5057249069213867,
|
| 82 |
"learning_rate": 0.0005,
|
| 83 |
-
"loss":
|
| 84 |
"step": 800
|
| 85 |
},
|
| 86 |
{
|
| 87 |
-
"epoch": 1.
|
| 88 |
-
"grad_norm": 0.5339187383651733,
|
| 89 |
"learning_rate": 0.0005,
|
| 90 |
-
"loss":
|
| 91 |
"step": 900
|
| 92 |
},
|
| 93 |
{
|
| 94 |
-
"epoch": 1.
|
| 95 |
-
"grad_norm": 0.5649825930595398,
|
| 96 |
"learning_rate": 0.0005,
|
| 97 |
-
"loss":
|
| 98 |
"step": 1000
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 2.0,
|
| 102 |
-
"eval_accuracy": 0.
|
| 103 |
-
"eval_loss":
|
| 104 |
-
"eval_runtime":
|
| 105 |
-
"eval_samples_per_second":
|
| 106 |
-
"eval_steps_per_second":
|
| 107 |
"step": 1058
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 2.0,
|
| 111 |
-
"eval_exact_match":
|
| 112 |
-
"eval_f1":
|
| 113 |
-
"eval_qa_bleu":
|
| 114 |
-
"eval_qa_exact_match": 0.
|
| 115 |
-
"eval_recite_bleu":
|
| 116 |
-
"eval_recite_exact_match": 0.
|
| 117 |
"step": 1058
|
| 118 |
},
|
| 119 |
{
|
| 120 |
-
"epoch": 2.
|
| 121 |
-
"grad_norm": 0.6042512655258179,
|
| 122 |
"learning_rate": 0.0005,
|
| 123 |
-
"loss":
|
| 124 |
"step": 1100
|
| 125 |
},
|
| 126 |
{
|
| 127 |
-
"epoch": 2.
|
| 128 |
-
"grad_norm": 965.2459106445312,
|
| 129 |
"learning_rate": 0.0005,
|
| 130 |
-
"loss": 8.
|
| 131 |
"step": 1200
|
| 132 |
},
|
| 133 |
{
|
| 134 |
-
"epoch": 2.
|
| 135 |
-
"grad_norm": 3.0128896236419678,
|
| 136 |
"learning_rate": 0.0005,
|
| 137 |
-
"loss":
|
| 138 |
"step": 1300
|
| 139 |
},
|
| 140 |
{
|
| 141 |
-
"epoch": 2.
|
| 142 |
-
"grad_norm": 108062179328.0,
|
| 143 |
"learning_rate": 0.0005,
|
| 144 |
-
"loss":
|
| 145 |
"step": 1400
|
| 146 |
},
|
| 147 |
{
|
| 148 |
-
"epoch": 2.
|
| 149 |
-
"grad_norm": 5002999889920.0,
|
| 150 |
"learning_rate": 0.0005,
|
| 151 |
-
"loss":
|
| 152 |
"step": 1500
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 3.0,
|
| 156 |
-
"eval_accuracy": 0.
|
| 157 |
-
"eval_loss":
|
| 158 |
-
"eval_runtime":
|
| 159 |
-
"eval_samples_per_second":
|
| 160 |
-
"eval_steps_per_second":
|
| 161 |
"step": 1587
|
| 162 |
},
|
| 163 |
{
|
|
@@ -171,54 +156,48 @@
|
|
| 171 |
"step": 1587
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch": 3.
|
| 175 |
-
"grad_norm": 62.19309616088867,
|
| 176 |
"learning_rate": 0.0005,
|
| 177 |
-
"loss":
|
| 178 |
"step": 1600
|
| 179 |
},
|
| 180 |
{
|
| 181 |
-
"epoch": 3.
|
| 182 |
-
"grad_norm": 4.109750747680664,
|
| 183 |
"learning_rate": 0.0005,
|
| 184 |
-
"loss":
|
| 185 |
"step": 1700
|
| 186 |
},
|
| 187 |
{
|
| 188 |
-
"epoch": 3.
|
| 189 |
-
"grad_norm": 7.620689392089844,
|
| 190 |
"learning_rate": 0.0005,
|
| 191 |
-
"loss":
|
| 192 |
"step": 1800
|
| 193 |
},
|
| 194 |
{
|
| 195 |
-
"epoch": 3.
|
| 196 |
-
"grad_norm": 33.092830657958984,
|
| 197 |
"learning_rate": 0.0005,
|
| 198 |
-
"loss": 7.
|
| 199 |
"step": 1900
|
| 200 |
},
|
| 201 |
{
|
| 202 |
-
"epoch": 3.
|
| 203 |
-
"grad_norm": 204828.203125,
|
| 204 |
"learning_rate": 0.0005,
|
| 205 |
-
"loss": 7.
|
| 206 |
"step": 2000
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch": 3.
|
| 210 |
-
"grad_norm": 229958729728.0,
|
| 211 |
"learning_rate": 0.0005,
|
| 212 |
-
"loss":
|
| 213 |
"step": 2100
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 4.0,
|
| 217 |
-
"eval_accuracy": 0.
|
| 218 |
-
"eval_loss":
|
| 219 |
-
"eval_runtime":
|
| 220 |
-
"eval_samples_per_second":
|
| 221 |
-
"eval_steps_per_second":
|
| 222 |
"step": 2116
|
| 223 |
},
|
| 224 |
{
|
|
@@ -232,47 +211,42 @@
|
|
| 232 |
"step": 2116
|
| 233 |
},
|
| 234 |
{
|
| 235 |
-
"epoch": 4.
|
| 236 |
-
"grad_norm": 17.098289489746094,
|
| 237 |
"learning_rate": 0.0005,
|
| 238 |
-
"loss":
|
| 239 |
"step": 2200
|
| 240 |
},
|
| 241 |
{
|
| 242 |
-
"epoch": 4.
|
| 243 |
-
"grad_norm": 1069.9039306640625,
|
| 244 |
"learning_rate": 0.0005,
|
| 245 |
-
"loss":
|
| 246 |
"step": 2300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
-
"epoch": 4.
|
| 250 |
-
"grad_norm": 7.694222927093506,
|
| 251 |
"learning_rate": 0.0005,
|
| 252 |
-
"loss": 7.
|
| 253 |
"step": 2400
|
| 254 |
},
|
| 255 |
{
|
| 256 |
-
"epoch": 4.
|
| 257 |
-
"grad_norm": Infinity,
|
| 258 |
"learning_rate": 0.0005,
|
| 259 |
-
"loss":
|
| 260 |
"step": 2500
|
| 261 |
},
|
| 262 |
{
|
| 263 |
-
"epoch": 4.
|
| 264 |
-
"grad_norm": 396939770200064.0,
|
| 265 |
"learning_rate": 0.0005,
|
| 266 |
-
"loss":
|
| 267 |
"step": 2600
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 5.0,
|
| 271 |
-
"eval_accuracy": 0.
|
| 272 |
-
"eval_loss":
|
| 273 |
-
"eval_runtime":
|
| 274 |
-
"eval_samples_per_second":
|
| 275 |
-
"eval_steps_per_second":
|
| 276 |
"step": 2645
|
| 277 |
},
|
| 278 |
{
|
|
@@ -286,47 +260,42 @@
|
|
| 286 |
"step": 2645
|
| 287 |
},
|
| 288 |
{
|
| 289 |
-
"epoch": 5.
|
| 290 |
-
"grad_norm": 3.166937635134374e+18,
|
| 291 |
"learning_rate": 0.0005,
|
| 292 |
-
"loss":
|
| 293 |
"step": 2700
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"epoch": 5.
|
| 297 |
-
"grad_norm": 1.8363503072027607e+18,
|
| 298 |
"learning_rate": 0.0005,
|
| 299 |
-
"loss":
|
| 300 |
"step": 2800
|
| 301 |
},
|
| 302 |
{
|
| 303 |
-
"epoch": 5.
|
| 304 |
-
"grad_norm": 113262112800768.0,
|
| 305 |
"learning_rate": 0.0005,
|
| 306 |
-
"loss":
|
| 307 |
"step": 2900
|
| 308 |
},
|
| 309 |
{
|
| 310 |
-
"epoch": 5.
|
| 311 |
-
"grad_norm": 1283434880.0,
|
| 312 |
"learning_rate": 0.0005,
|
| 313 |
-
"loss":
|
| 314 |
"step": 3000
|
| 315 |
},
|
| 316 |
{
|
| 317 |
-
"epoch": 5.
|
| 318 |
-
"grad_norm": 1221250.5,
|
| 319 |
"learning_rate": 0.0005,
|
| 320 |
-
"loss":
|
| 321 |
"step": 3100
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 6.0,
|
| 325 |
-
"eval_accuracy": 0.
|
| 326 |
-
"eval_loss":
|
| 327 |
-
"eval_runtime":
|
| 328 |
-
"eval_samples_per_second":
|
| 329 |
-
"eval_steps_per_second":
|
| 330 |
"step": 3174
|
| 331 |
},
|
| 332 |
{
|
|
@@ -340,54 +309,48 @@
|
|
| 340 |
"step": 3174
|
| 341 |
},
|
| 342 |
{
|
| 343 |
-
"epoch": 6.
|
| 344 |
-
"grad_norm": 2019.499267578125,
|
| 345 |
"learning_rate": 0.0005,
|
| 346 |
-
"loss":
|
| 347 |
"step": 3200
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch": 6.
|
| 351 |
-
"grad_norm": 164.9034881591797,
|
| 352 |
"learning_rate": 0.0005,
|
| 353 |
-
"loss":
|
| 354 |
"step": 3300
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch": 6.
|
| 358 |
-
"grad_norm": 434.93157958984375,
|
| 359 |
"learning_rate": 0.0005,
|
| 360 |
-
"loss": 7.
|
| 361 |
"step": 3400
|
| 362 |
},
|
| 363 |
{
|
| 364 |
-
"epoch": 6.
|
| 365 |
-
"grad_norm": 1324.7332763671875,
|
| 366 |
"learning_rate": 0.0005,
|
| 367 |
-
"loss": 7.
|
| 368 |
"step": 3500
|
| 369 |
},
|
| 370 |
{
|
| 371 |
-
"epoch": 6.
|
| 372 |
-
"grad_norm": 52531002605568.0,
|
| 373 |
"learning_rate": 0.0005,
|
| 374 |
-
"loss":
|
| 375 |
"step": 3600
|
| 376 |
},
|
| 377 |
{
|
| 378 |
-
"epoch": 6.
|
| 379 |
-
"grad_norm": Infinity,
|
| 380 |
"learning_rate": 0.0005,
|
| 381 |
-
"loss":
|
| 382 |
"step": 3700
|
| 383 |
},
|
| 384 |
{
|
| 385 |
"epoch": 7.0,
|
| 386 |
-
"eval_accuracy": 0.
|
| 387 |
-
"eval_loss":
|
| 388 |
-
"eval_runtime":
|
| 389 |
-
"eval_samples_per_second":
|
| 390 |
-
"eval_steps_per_second":
|
| 391 |
"step": 3703
|
| 392 |
},
|
| 393 |
{
|
|
@@ -401,47 +364,42 @@
|
|
| 401 |
"step": 3703
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 7.
|
| 405 |
-
"grad_norm": Infinity,
|
| 406 |
"learning_rate": 0.0005,
|
| 407 |
-
"loss":
|
| 408 |
"step": 3800
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch": 7.
|
| 412 |
-
"grad_norm": Infinity,
|
| 413 |
"learning_rate": 0.0005,
|
| 414 |
-
"loss":
|
| 415 |
"step": 3900
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch": 7.
|
| 419 |
-
"grad_norm": Infinity,
|
| 420 |
"learning_rate": 0.0005,
|
| 421 |
-
"loss":
|
| 422 |
"step": 4000
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch": 7.
|
| 426 |
-
"grad_norm": Infinity,
|
| 427 |
"learning_rate": 0.0005,
|
| 428 |
-
"loss":
|
| 429 |
"step": 4100
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 7.
|
| 433 |
-
"grad_norm": NaN,
|
| 434 |
"learning_rate": 0.0005,
|
| 435 |
-
"loss":
|
| 436 |
"step": 4200
|
| 437 |
},
|
| 438 |
{
|
| 439 |
"epoch": 8.0,
|
| 440 |
-
"eval_accuracy": 0.
|
| 441 |
-
"eval_loss":
|
| 442 |
-
"eval_runtime":
|
| 443 |
-
"eval_samples_per_second":
|
| 444 |
-
"eval_steps_per_second":
|
| 445 |
"step": 4232
|
| 446 |
},
|
| 447 |
{
|
|
@@ -455,47 +413,42 @@
|
|
| 455 |
"step": 4232
|
| 456 |
},
|
| 457 |
{
|
| 458 |
-
"epoch": 8.
|
| 459 |
-
"grad_norm": NaN,
|
| 460 |
"learning_rate": 0.0005,
|
| 461 |
-
"loss":
|
| 462 |
"step": 4300
|
| 463 |
},
|
| 464 |
{
|
| 465 |
-
"epoch": 8.
|
| 466 |
-
"grad_norm": NaN,
|
| 467 |
"learning_rate": 0.0005,
|
| 468 |
-
"loss":
|
| 469 |
"step": 4400
|
| 470 |
},
|
| 471 |
{
|
| 472 |
-
"epoch": 8.
|
| 473 |
-
"grad_norm": NaN,
|
| 474 |
"learning_rate": 0.0005,
|
| 475 |
-
"loss":
|
| 476 |
"step": 4500
|
| 477 |
},
|
| 478 |
{
|
| 479 |
-
"epoch": 8.
|
| 480 |
-
"grad_norm": NaN,
|
| 481 |
"learning_rate": 0.0005,
|
| 482 |
-
"loss":
|
| 483 |
"step": 4600
|
| 484 |
},
|
| 485 |
{
|
| 486 |
-
"epoch": 8.
|
| 487 |
-
"grad_norm": NaN,
|
| 488 |
"learning_rate": 0.0005,
|
| 489 |
-
"loss":
|
| 490 |
"step": 4700
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 9.0,
|
| 494 |
-
"eval_accuracy": 0.
|
| 495 |
-
"eval_loss":
|
| 496 |
-
"eval_runtime":
|
| 497 |
-
"eval_samples_per_second":
|
| 498 |
-
"eval_steps_per_second":
|
| 499 |
"step": 4761
|
| 500 |
},
|
| 501 |
{
|
|
@@ -509,47 +462,42 @@
|
|
| 509 |
"step": 4761
|
| 510 |
},
|
| 511 |
{
|
| 512 |
-
"epoch": 9.
|
| 513 |
-
"grad_norm": NaN,
|
| 514 |
"learning_rate": 0.0005,
|
| 515 |
-
"loss":
|
| 516 |
"step": 4800
|
| 517 |
},
|
| 518 |
{
|
| 519 |
-
"epoch": 9.
|
| 520 |
-
"grad_norm": NaN,
|
| 521 |
"learning_rate": 0.0005,
|
| 522 |
-
"loss":
|
| 523 |
"step": 4900
|
| 524 |
},
|
| 525 |
{
|
| 526 |
-
"epoch": 9.
|
| 527 |
-
"grad_norm": NaN,
|
| 528 |
"learning_rate": 0.0005,
|
| 529 |
-
"loss":
|
| 530 |
"step": 5000
|
| 531 |
},
|
| 532 |
{
|
| 533 |
-
"epoch": 9.
|
| 534 |
-
"grad_norm": NaN,
|
| 535 |
"learning_rate": 0.0005,
|
| 536 |
-
"loss":
|
| 537 |
"step": 5100
|
| 538 |
},
|
| 539 |
{
|
| 540 |
-
"epoch": 9.
|
| 541 |
-
"grad_norm": NaN,
|
| 542 |
"learning_rate": 0.0005,
|
| 543 |
-
"loss":
|
| 544 |
"step": 5200
|
| 545 |
},
|
| 546 |
{
|
| 547 |
"epoch": 10.0,
|
| 548 |
-
"eval_accuracy": 0.
|
| 549 |
-
"eval_loss":
|
| 550 |
-
"eval_runtime":
|
| 551 |
-
"eval_samples_per_second":
|
| 552 |
-
"eval_steps_per_second":
|
| 553 |
"step": 5290
|
| 554 |
},
|
| 555 |
{
|
|
@@ -563,54 +511,48 @@
|
|
| 563 |
"step": 5290
|
| 564 |
},
|
| 565 |
{
|
| 566 |
-
"epoch": 10.
|
| 567 |
-
"grad_norm": NaN,
|
| 568 |
"learning_rate": 0.0005,
|
| 569 |
-
"loss":
|
| 570 |
"step": 5300
|
| 571 |
},
|
| 572 |
{
|
| 573 |
-
"epoch": 10.
|
| 574 |
-
"grad_norm": NaN,
|
| 575 |
"learning_rate": 0.0005,
|
| 576 |
-
"loss":
|
| 577 |
"step": 5400
|
| 578 |
},
|
| 579 |
{
|
| 580 |
-
"epoch": 10.
|
| 581 |
-
"grad_norm": NaN,
|
| 582 |
"learning_rate": 0.0005,
|
| 583 |
-
"loss":
|
| 584 |
"step": 5500
|
| 585 |
},
|
| 586 |
{
|
| 587 |
-
"epoch": 10.
|
| 588 |
-
"grad_norm": NaN,
|
| 589 |
"learning_rate": 0.0005,
|
| 590 |
-
"loss":
|
| 591 |
"step": 5600
|
| 592 |
},
|
| 593 |
{
|
| 594 |
-
"epoch": 10.
|
| 595 |
-
"grad_norm": NaN,
|
| 596 |
"learning_rate": 0.0005,
|
| 597 |
-
"loss":
|
| 598 |
"step": 5700
|
| 599 |
},
|
| 600 |
{
|
| 601 |
-
"epoch": 10.
|
| 602 |
-
"grad_norm": NaN,
|
| 603 |
"learning_rate": 0.0005,
|
| 604 |
-
"loss":
|
| 605 |
"step": 5800
|
| 606 |
},
|
| 607 |
{
|
| 608 |
"epoch": 11.0,
|
| 609 |
-
"eval_accuracy": 0.
|
| 610 |
-
"eval_loss":
|
| 611 |
-
"eval_runtime":
|
| 612 |
-
"eval_samples_per_second":
|
| 613 |
-
"eval_steps_per_second":
|
| 614 |
"step": 5819
|
| 615 |
},
|
| 616 |
{
|
|
@@ -624,47 +566,42 @@
|
|
| 624 |
"step": 5819
|
| 625 |
},
|
| 626 |
{
|
| 627 |
-
"epoch": 11.
|
| 628 |
-
"grad_norm": NaN,
|
| 629 |
"learning_rate": 0.0005,
|
| 630 |
-
"loss":
|
| 631 |
"step": 5900
|
| 632 |
},
|
| 633 |
{
|
| 634 |
-
"epoch": 11.
|
| 635 |
-
"grad_norm": NaN,
|
| 636 |
"learning_rate": 0.0005,
|
| 637 |
-
"loss":
|
| 638 |
"step": 6000
|
| 639 |
},
|
| 640 |
{
|
| 641 |
-
"epoch": 11.
|
| 642 |
-
"grad_norm": NaN,
|
| 643 |
"learning_rate": 0.0005,
|
| 644 |
-
"loss":
|
| 645 |
"step": 6100
|
| 646 |
},
|
| 647 |
{
|
| 648 |
-
"epoch": 11.
|
| 649 |
-
"grad_norm": NaN,
|
| 650 |
"learning_rate": 0.0005,
|
| 651 |
-
"loss":
|
| 652 |
"step": 6200
|
| 653 |
},
|
| 654 |
{
|
| 655 |
-
"epoch": 11.
|
| 656 |
-
"grad_norm": NaN,
|
| 657 |
"learning_rate": 0.0005,
|
| 658 |
-
"loss":
|
| 659 |
"step": 6300
|
| 660 |
},
|
| 661 |
{
|
| 662 |
"epoch": 12.0,
|
| 663 |
-
"eval_accuracy": 0.
|
| 664 |
-
"eval_loss":
|
| 665 |
-
"eval_runtime":
|
| 666 |
-
"eval_samples_per_second":
|
| 667 |
-
"eval_steps_per_second":
|
| 668 |
"step": 6348
|
| 669 |
},
|
| 670 |
{
|
|
@@ -678,47 +615,42 @@
|
|
| 678 |
"step": 6348
|
| 679 |
},
|
| 680 |
{
|
| 681 |
-
"epoch": 12.
|
| 682 |
-
"grad_norm": NaN,
|
| 683 |
"learning_rate": 0.0005,
|
| 684 |
-
"loss":
|
| 685 |
"step": 6400
|
| 686 |
},
|
| 687 |
{
|
| 688 |
-
"epoch": 12.
|
| 689 |
-
"grad_norm": NaN,
|
| 690 |
"learning_rate": 0.0005,
|
| 691 |
-
"loss":
|
| 692 |
"step": 6500
|
| 693 |
},
|
| 694 |
{
|
| 695 |
-
"epoch": 12.
|
| 696 |
-
"grad_norm": NaN,
|
| 697 |
"learning_rate": 0.0005,
|
| 698 |
-
"loss":
|
| 699 |
"step": 6600
|
| 700 |
},
|
| 701 |
{
|
| 702 |
-
"epoch": 12.
|
| 703 |
-
"grad_norm": NaN,
|
| 704 |
"learning_rate": 0.0005,
|
| 705 |
-
"loss":
|
| 706 |
"step": 6700
|
| 707 |
},
|
| 708 |
{
|
| 709 |
-
"epoch": 12.
|
| 710 |
-
"grad_norm": NaN,
|
| 711 |
"learning_rate": 0.0005,
|
| 712 |
-
"loss":
|
| 713 |
"step": 6800
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 13.0,
|
| 717 |
-
"eval_accuracy": 0.
|
| 718 |
-
"eval_loss":
|
| 719 |
-
"eval_runtime":
|
| 720 |
-
"eval_samples_per_second":
|
| 721 |
-
"eval_steps_per_second":
|
| 722 |
"step": 6877
|
| 723 |
},
|
| 724 |
{
|
|
@@ -732,54 +664,48 @@
|
|
| 732 |
"step": 6877
|
| 733 |
},
|
| 734 |
{
|
| 735 |
-
"epoch": 13.
|
| 736 |
-
"grad_norm": NaN,
|
| 737 |
"learning_rate": 0.0005,
|
| 738 |
-
"loss":
|
| 739 |
"step": 6900
|
| 740 |
},
|
| 741 |
{
|
| 742 |
-
"epoch": 13.
|
| 743 |
-
"grad_norm": NaN,
|
| 744 |
"learning_rate": 0.0005,
|
| 745 |
-
"loss":
|
| 746 |
"step": 7000
|
| 747 |
},
|
| 748 |
{
|
| 749 |
-
"epoch": 13.
|
| 750 |
-
"grad_norm": NaN,
|
| 751 |
"learning_rate": 0.0005,
|
| 752 |
-
"loss":
|
| 753 |
"step": 7100
|
| 754 |
},
|
| 755 |
{
|
| 756 |
-
"epoch": 13.
|
| 757 |
-
"grad_norm": NaN,
|
| 758 |
"learning_rate": 0.0005,
|
| 759 |
-
"loss":
|
| 760 |
"step": 7200
|
| 761 |
},
|
| 762 |
{
|
| 763 |
-
"epoch": 13.
|
| 764 |
-
"grad_norm": NaN,
|
| 765 |
"learning_rate": 0.0005,
|
| 766 |
-
"loss":
|
| 767 |
"step": 7300
|
| 768 |
},
|
| 769 |
{
|
| 770 |
-
"epoch": 13.
|
| 771 |
-
"grad_norm": NaN,
|
| 772 |
"learning_rate": 0.0005,
|
| 773 |
-
"loss":
|
| 774 |
"step": 7400
|
| 775 |
},
|
| 776 |
{
|
| 777 |
"epoch": 14.0,
|
| 778 |
-
"eval_accuracy": 0.
|
| 779 |
-
"eval_loss":
|
| 780 |
-
"eval_runtime":
|
| 781 |
-
"eval_samples_per_second":
|
| 782 |
-
"eval_steps_per_second":
|
| 783 |
"step": 7406
|
| 784 |
},
|
| 785 |
{
|
|
@@ -793,47 +719,42 @@
|
|
| 793 |
"step": 7406
|
| 794 |
},
|
| 795 |
{
|
| 796 |
-
"epoch": 14.
|
| 797 |
-
"grad_norm": NaN,
|
| 798 |
"learning_rate": 0.0005,
|
| 799 |
-
"loss":
|
| 800 |
"step": 7500
|
| 801 |
},
|
| 802 |
{
|
| 803 |
-
"epoch": 14.
|
| 804 |
-
"grad_norm": NaN,
|
| 805 |
"learning_rate": 0.0005,
|
| 806 |
-
"loss":
|
| 807 |
"step": 7600
|
| 808 |
},
|
| 809 |
{
|
| 810 |
-
"epoch": 14.
|
| 811 |
-
"grad_norm": NaN,
|
| 812 |
"learning_rate": 0.0005,
|
| 813 |
-
"loss":
|
| 814 |
"step": 7700
|
| 815 |
},
|
| 816 |
{
|
| 817 |
-
"epoch": 14.
|
| 818 |
-
"grad_norm": NaN,
|
| 819 |
"learning_rate": 0.0005,
|
| 820 |
-
"loss":
|
| 821 |
"step": 7800
|
| 822 |
},
|
| 823 |
{
|
| 824 |
-
"epoch": 14.
|
| 825 |
-
"grad_norm": NaN,
|
| 826 |
"learning_rate": 0.0005,
|
| 827 |
-
"loss":
|
| 828 |
"step": 7900
|
| 829 |
},
|
| 830 |
{
|
| 831 |
"epoch": 15.0,
|
| 832 |
-
"eval_accuracy": 0.
|
| 833 |
-
"eval_loss":
|
| 834 |
-
"eval_runtime":
|
| 835 |
-
"eval_samples_per_second":
|
| 836 |
-
"eval_steps_per_second":
|
| 837 |
"step": 7935
|
| 838 |
},
|
| 839 |
{
|
|
@@ -847,47 +768,42 @@
|
|
| 847 |
"step": 7935
|
| 848 |
},
|
| 849 |
{
|
| 850 |
-
"epoch": 15.
|
| 851 |
-
"grad_norm": NaN,
|
| 852 |
"learning_rate": 0.0005,
|
| 853 |
-
"loss":
|
| 854 |
"step": 8000
|
| 855 |
},
|
| 856 |
{
|
| 857 |
-
"epoch": 15.
|
| 858 |
-
"grad_norm": NaN,
|
| 859 |
"learning_rate": 0.0005,
|
| 860 |
-
"loss":
|
| 861 |
"step": 8100
|
| 862 |
},
|
| 863 |
{
|
| 864 |
-
"epoch": 15.
|
| 865 |
-
"grad_norm": NaN,
|
| 866 |
"learning_rate": 0.0005,
|
| 867 |
-
"loss":
|
| 868 |
"step": 8200
|
| 869 |
},
|
| 870 |
{
|
| 871 |
-
"epoch": 15.
|
| 872 |
-
"grad_norm": NaN,
|
| 873 |
"learning_rate": 0.0005,
|
| 874 |
-
"loss":
|
| 875 |
"step": 8300
|
| 876 |
},
|
| 877 |
{
|
| 878 |
-
"epoch": 15.
|
| 879 |
-
"grad_norm": NaN,
|
| 880 |
"learning_rate": 0.0005,
|
| 881 |
-
"loss":
|
| 882 |
"step": 8400
|
| 883 |
},
|
| 884 |
{
|
| 885 |
"epoch": 16.0,
|
| 886 |
-
"eval_accuracy": 0.
|
| 887 |
-
"eval_loss":
|
| 888 |
-
"eval_runtime":
|
| 889 |
-
"eval_samples_per_second":
|
| 890 |
-
"eval_steps_per_second":
|
| 891 |
"step": 8464
|
| 892 |
},
|
| 893 |
{
|
|
@@ -901,47 +817,42 @@
|
|
| 901 |
"step": 8464
|
| 902 |
},
|
| 903 |
{
|
| 904 |
-
"epoch": 16.
|
| 905 |
-
"grad_norm": NaN,
|
| 906 |
"learning_rate": 0.0005,
|
| 907 |
-
"loss":
|
| 908 |
"step": 8500
|
| 909 |
},
|
| 910 |
{
|
| 911 |
-
"epoch": 16.
|
| 912 |
-
"grad_norm": NaN,
|
| 913 |
"learning_rate": 0.0005,
|
| 914 |
-
"loss":
|
| 915 |
"step": 8600
|
| 916 |
},
|
| 917 |
{
|
| 918 |
-
"epoch": 16.
|
| 919 |
-
"grad_norm": NaN,
|
| 920 |
"learning_rate": 0.0005,
|
| 921 |
-
"loss":
|
| 922 |
"step": 8700
|
| 923 |
},
|
| 924 |
{
|
| 925 |
-
"epoch": 16.
|
| 926 |
-
"grad_norm": NaN,
|
| 927 |
"learning_rate": 0.0005,
|
| 928 |
-
"loss":
|
| 929 |
"step": 8800
|
| 930 |
},
|
| 931 |
{
|
| 932 |
-
"epoch": 16.
|
| 933 |
-
"grad_norm": NaN,
|
| 934 |
"learning_rate": 0.0005,
|
| 935 |
-
"loss":
|
| 936 |
"step": 8900
|
| 937 |
},
|
| 938 |
{
|
| 939 |
"epoch": 17.0,
|
| 940 |
-
"eval_accuracy": 0.
|
| 941 |
-
"eval_loss":
|
| 942 |
-
"eval_runtime":
|
| 943 |
-
"eval_samples_per_second":
|
| 944 |
-
"eval_steps_per_second":
|
| 945 |
"step": 8993
|
| 946 |
},
|
| 947 |
{
|
|
@@ -955,54 +866,48 @@
|
|
| 955 |
"step": 8993
|
| 956 |
},
|
| 957 |
{
|
| 958 |
-
"epoch": 17.
|
| 959 |
-
"grad_norm": NaN,
|
| 960 |
"learning_rate": 0.0005,
|
| 961 |
-
"loss":
|
| 962 |
"step": 9000
|
| 963 |
},
|
| 964 |
{
|
| 965 |
-
"epoch": 17.
|
| 966 |
-
"grad_norm": NaN,
|
| 967 |
"learning_rate": 0.0005,
|
| 968 |
-
"loss":
|
| 969 |
"step": 9100
|
| 970 |
},
|
| 971 |
{
|
| 972 |
-
"epoch": 17.
|
| 973 |
-
"grad_norm": NaN,
|
| 974 |
"learning_rate": 0.0005,
|
| 975 |
-
"loss":
|
| 976 |
"step": 9200
|
| 977 |
},
|
| 978 |
{
|
| 979 |
-
"epoch": 17.
|
| 980 |
-
"grad_norm": NaN,
|
| 981 |
"learning_rate": 0.0005,
|
| 982 |
-
"loss":
|
| 983 |
"step": 9300
|
| 984 |
},
|
| 985 |
{
|
| 986 |
-
"epoch": 17.
|
| 987 |
-
"grad_norm": NaN,
|
| 988 |
"learning_rate": 0.0005,
|
| 989 |
-
"loss":
|
| 990 |
"step": 9400
|
| 991 |
},
|
| 992 |
{
|
| 993 |
-
"epoch": 17.
|
| 994 |
-
"grad_norm": NaN,
|
| 995 |
"learning_rate": 0.0005,
|
| 996 |
-
"loss":
|
| 997 |
"step": 9500
|
| 998 |
},
|
| 999 |
{
|
| 1000 |
"epoch": 18.0,
|
| 1001 |
-
"eval_accuracy": 0.
|
| 1002 |
-
"eval_loss":
|
| 1003 |
-
"eval_runtime":
|
| 1004 |
-
"eval_samples_per_second":
|
| 1005 |
-
"eval_steps_per_second":
|
| 1006 |
"step": 9522
|
| 1007 |
},
|
| 1008 |
{
|
|
@@ -1016,47 +921,42 @@
|
|
| 1016 |
"step": 9522
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
-
"epoch": 18.
|
| 1020 |
-
"grad_norm": NaN,
|
| 1021 |
"learning_rate": 0.0005,
|
| 1022 |
-
"loss":
|
| 1023 |
"step": 9600
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
-
"epoch": 18.
|
| 1027 |
-
"grad_norm": NaN,
|
| 1028 |
"learning_rate": 0.0005,
|
| 1029 |
-
"loss":
|
| 1030 |
"step": 9700
|
| 1031 |
},
|
| 1032 |
{
|
| 1033 |
-
"epoch": 18.
|
| 1034 |
-
"grad_norm": NaN,
|
| 1035 |
"learning_rate": 0.0005,
|
| 1036 |
-
"loss":
|
| 1037 |
"step": 9800
|
| 1038 |
},
|
| 1039 |
{
|
| 1040 |
-
"epoch": 18.
|
| 1041 |
-
"grad_norm": NaN,
|
| 1042 |
"learning_rate": 0.0005,
|
| 1043 |
-
"loss":
|
| 1044 |
"step": 9900
|
| 1045 |
},
|
| 1046 |
{
|
| 1047 |
-
"epoch": 18.
|
| 1048 |
-
"grad_norm": NaN,
|
| 1049 |
"learning_rate": 0.0005,
|
| 1050 |
-
"loss":
|
| 1051 |
"step": 10000
|
| 1052 |
},
|
| 1053 |
{
|
| 1054 |
"epoch": 19.0,
|
| 1055 |
-
"eval_accuracy": 0.
|
| 1056 |
-
"eval_loss":
|
| 1057 |
-
"eval_runtime":
|
| 1058 |
-
"eval_samples_per_second":
|
| 1059 |
-
"eval_steps_per_second":
|
| 1060 |
"step": 10051
|
| 1061 |
},
|
| 1062 |
{
|
|
@@ -1070,47 +970,42 @@
|
|
| 1070 |
"step": 10051
|
| 1071 |
},
|
| 1072 |
{
|
| 1073 |
-
"epoch": 19.
|
| 1074 |
-
"grad_norm": NaN,
|
| 1075 |
"learning_rate": 0.0005,
|
| 1076 |
-
"loss":
|
| 1077 |
"step": 10100
|
| 1078 |
},
|
| 1079 |
{
|
| 1080 |
-
"epoch": 19.
|
| 1081 |
-
"grad_norm": NaN,
|
| 1082 |
"learning_rate": 0.0005,
|
| 1083 |
-
"loss":
|
| 1084 |
"step": 10200
|
| 1085 |
},
|
| 1086 |
{
|
| 1087 |
-
"epoch": 19.
|
| 1088 |
-
"grad_norm": NaN,
|
| 1089 |
"learning_rate": 0.0005,
|
| 1090 |
-
"loss":
|
| 1091 |
"step": 10300
|
| 1092 |
},
|
| 1093 |
{
|
| 1094 |
-
"epoch": 19.
|
| 1095 |
-
"grad_norm": NaN,
|
| 1096 |
"learning_rate": 0.0005,
|
| 1097 |
-
"loss":
|
| 1098 |
"step": 10400
|
| 1099 |
},
|
| 1100 |
{
|
| 1101 |
-
"epoch": 19.
|
| 1102 |
-
"grad_norm": NaN,
|
| 1103 |
"learning_rate": 0.0005,
|
| 1104 |
-
"loss":
|
| 1105 |
"step": 10500
|
| 1106 |
},
|
| 1107 |
{
|
| 1108 |
"epoch": 20.0,
|
| 1109 |
-
"eval_accuracy": 0.
|
| 1110 |
-
"eval_loss":
|
| 1111 |
-
"eval_runtime": 6.
|
| 1112 |
-
"eval_samples_per_second":
|
| 1113 |
-
"eval_steps_per_second": 9.
|
| 1114 |
"step": 10580
|
| 1115 |
},
|
| 1116 |
{
|
|
@@ -1124,54 +1019,48 @@
|
|
| 1124 |
"step": 10580
|
| 1125 |
},
|
| 1126 |
{
|
| 1127 |
-
"epoch": 20.
|
| 1128 |
-
"grad_norm": NaN,
|
| 1129 |
"learning_rate": 0.0005,
|
| 1130 |
-
"loss":
|
| 1131 |
"step": 10600
|
| 1132 |
},
|
| 1133 |
{
|
| 1134 |
-
"epoch": 20.
|
| 1135 |
-
"grad_norm": NaN,
|
| 1136 |
"learning_rate": 0.0005,
|
| 1137 |
-
"loss":
|
| 1138 |
"step": 10700
|
| 1139 |
},
|
| 1140 |
{
|
| 1141 |
-
"epoch": 20.
|
| 1142 |
-
"grad_norm": NaN,
|
| 1143 |
"learning_rate": 0.0005,
|
| 1144 |
-
"loss":
|
| 1145 |
"step": 10800
|
| 1146 |
},
|
| 1147 |
{
|
| 1148 |
-
"epoch": 20.
|
| 1149 |
-
"grad_norm": NaN,
|
| 1150 |
"learning_rate": 0.0005,
|
| 1151 |
-
"loss":
|
| 1152 |
"step": 10900
|
| 1153 |
},
|
| 1154 |
{
|
| 1155 |
-
"epoch": 20.
|
| 1156 |
-
"grad_norm": NaN,
|
| 1157 |
"learning_rate": 0.0005,
|
| 1158 |
-
"loss":
|
| 1159 |
"step": 11000
|
| 1160 |
},
|
| 1161 |
{
|
| 1162 |
-
"epoch": 20.
|
| 1163 |
-
"grad_norm": NaN,
|
| 1164 |
"learning_rate": 0.0005,
|
| 1165 |
-
"loss":
|
| 1166 |
"step": 11100
|
| 1167 |
},
|
| 1168 |
{
|
| 1169 |
"epoch": 21.0,
|
| 1170 |
-
"eval_accuracy": 0.
|
| 1171 |
-
"eval_loss":
|
| 1172 |
-
"eval_runtime":
|
| 1173 |
-
"eval_samples_per_second":
|
| 1174 |
-
"eval_steps_per_second":
|
| 1175 |
"step": 11109
|
| 1176 |
},
|
| 1177 |
{
|
|
@@ -1187,23 +1076,9 @@
|
|
| 1187 |
],
|
| 1188 |
"logging_steps": 100,
|
| 1189 |
"max_steps": 26450,
|
| 1190 |
-
"num_input_tokens_seen": 0,
|
| 1191 |
"num_train_epochs": 50,
|
| 1192 |
"save_steps": 500,
|
| 1193 |
-
"
|
| 1194 |
-
"TrainerControl": {
|
| 1195 |
-
"args": {
|
| 1196 |
-
"should_epoch_stop": false,
|
| 1197 |
-
"should_evaluate": false,
|
| 1198 |
-
"should_log": false,
|
| 1199 |
-
"should_save": true,
|
| 1200 |
-
"should_training_stop": false
|
| 1201 |
-
},
|
| 1202 |
-
"attributes": {}
|
| 1203 |
-
}
|
| 1204 |
-
},
|
| 1205 |
-
"total_flos": 1.0124538570006856e+18,
|
| 1206 |
-
"train_batch_size": 1,
|
| 1207 |
"trial_name": null,
|
| 1208 |
"trial_params": null
|
| 1209 |
}
|
|
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.19,
|
|
|
|
| 13 |
"learning_rate": 0.0005,
|
| 14 |
+
"loss": 1.3631,
|
| 15 |
"step": 100
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.38,
|
|
|
|
| 19 |
"learning_rate": 0.0005,
|
| 20 |
+
"loss": 1.294,
|
| 21 |
"step": 200
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.57,
|
|
|
|
| 25 |
"learning_rate": 0.0005,
|
| 26 |
+
"loss": 1.302,
|
| 27 |
"step": 300
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.76,
|
|
|
|
| 31 |
"learning_rate": 0.0005,
|
| 32 |
+
"loss": 1.2626,
|
| 33 |
"step": 400
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.95,
|
|
|
|
| 37 |
"learning_rate": 0.0005,
|
| 38 |
+
"loss": 1.2156,
|
| 39 |
"step": 500
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 1.0,
|
| 43 |
+
"eval_accuracy": 0.6911294117647059,
|
| 44 |
+
"eval_loss": 1.012805700302124,
|
| 45 |
+
"eval_runtime": 7.0918,
|
| 46 |
+
"eval_samples_per_second": 70.504,
|
| 47 |
+
"eval_steps_per_second": 8.883,
|
| 48 |
"step": 529
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.0,
|
| 52 |
+
"eval_exact_match": 27.6,
|
| 53 |
+
"eval_f1": 35.68145743145743,
|
| 54 |
+
"eval_qa_bleu": 19.064111689202868,
|
| 55 |
+
"eval_qa_exact_match": 0.216,
|
| 56 |
+
"eval_recite_bleu": 17.898525975872754,
|
| 57 |
+
"eval_recite_exact_match": 0.006,
|
| 58 |
"step": 529
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 1.13,
|
|
|
|
| 62 |
"learning_rate": 0.0005,
|
| 63 |
+
"loss": 1.0305,
|
| 64 |
"step": 600
|
| 65 |
},
|
| 66 |
{
|
| 67 |
+
"epoch": 1.32,
|
|
|
|
| 68 |
"learning_rate": 0.0005,
|
| 69 |
+
"loss": 0.9199,
|
| 70 |
"step": 700
|
| 71 |
},
|
| 72 |
{
|
| 73 |
+
"epoch": 1.51,
|
|
|
|
| 74 |
"learning_rate": 0.0005,
|
| 75 |
+
"loss": 0.9086,
|
| 76 |
"step": 800
|
| 77 |
},
|
| 78 |
{
|
| 79 |
+
"epoch": 1.7,
|
|
|
|
| 80 |
"learning_rate": 0.0005,
|
| 81 |
+
"loss": 2.9245,
|
| 82 |
"step": 900
|
| 83 |
},
|
| 84 |
{
|
| 85 |
+
"epoch": 1.89,
|
|
|
|
| 86 |
"learning_rate": 0.0005,
|
| 87 |
+
"loss": 4.9199,
|
| 88 |
"step": 1000
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 2.0,
|
| 92 |
+
"eval_accuracy": 0.22922352941176471,
|
| 93 |
+
"eval_loss": 7.1224870681762695,
|
| 94 |
+
"eval_runtime": 7.7159,
|
| 95 |
+
"eval_samples_per_second": 64.801,
|
| 96 |
+
"eval_steps_per_second": 8.165,
|
| 97 |
"step": 1058
|
| 98 |
},
|
| 99 |
{
|
| 100 |
"epoch": 2.0,
|
| 101 |
+
"eval_exact_match": 0.0,
|
| 102 |
+
"eval_f1": 0.0,
|
| 103 |
+
"eval_qa_bleu": 0.0,
|
| 104 |
+
"eval_qa_exact_match": 0.0,
|
| 105 |
+
"eval_recite_bleu": 0.0,
|
| 106 |
+
"eval_recite_exact_match": 0.0,
|
| 107 |
"step": 1058
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 2.08,
|
|
|
|
| 111 |
"learning_rate": 0.0005,
|
| 112 |
+
"loss": 7.4349,
|
| 113 |
"step": 1100
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"epoch": 2.27,
|
|
|
|
| 117 |
"learning_rate": 0.0005,
|
| 118 |
+
"loss": 8.0921,
|
| 119 |
"step": 1200
|
| 120 |
},
|
| 121 |
{
|
| 122 |
+
"epoch": 2.46,
|
|
|
|
| 123 |
"learning_rate": 0.0005,
|
| 124 |
+
"loss": 7.889,
|
| 125 |
"step": 1300
|
| 126 |
},
|
| 127 |
{
|
| 128 |
+
"epoch": 2.65,
|
|
|
|
| 129 |
"learning_rate": 0.0005,
|
| 130 |
+
"loss": 7.2559,
|
| 131 |
"step": 1400
|
| 132 |
},
|
| 133 |
{
|
| 134 |
+
"epoch": 2.84,
|
|
|
|
| 135 |
"learning_rate": 0.0005,
|
| 136 |
+
"loss": 7.3563,
|
| 137 |
"step": 1500
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 3.0,
|
| 141 |
+
"eval_accuracy": 0.24235294117647058,
|
| 142 |
+
"eval_loss": 7.308440685272217,
|
| 143 |
+
"eval_runtime": 7.7087,
|
| 144 |
+
"eval_samples_per_second": 64.862,
|
| 145 |
+
"eval_steps_per_second": 8.173,
|
| 146 |
"step": 1587
|
| 147 |
},
|
| 148 |
{
|
|
|
|
| 156 |
"step": 1587
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 3.02,
|
|
|
|
| 160 |
"learning_rate": 0.0005,
|
| 161 |
+
"loss": 7.2721,
|
| 162 |
"step": 1600
|
| 163 |
},
|
| 164 |
{
|
| 165 |
+
"epoch": 3.21,
|
|
|
|
| 166 |
"learning_rate": 0.0005,
|
| 167 |
+
"loss": 7.2743,
|
| 168 |
"step": 1700
|
| 169 |
},
|
| 170 |
{
|
| 171 |
+
"epoch": 3.4,
|
|
|
|
| 172 |
"learning_rate": 0.0005,
|
| 173 |
+
"loss": 7.3151,
|
| 174 |
"step": 1800
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 3.59,
|
|
|
|
| 178 |
"learning_rate": 0.0005,
|
| 179 |
+
"loss": 7.3366,
|
| 180 |
"step": 1900
|
| 181 |
},
|
| 182 |
{
|
| 183 |
+
"epoch": 3.78,
|
|
|
|
| 184 |
"learning_rate": 0.0005,
|
| 185 |
+
"loss": 7.3294,
|
| 186 |
"step": 2000
|
| 187 |
},
|
| 188 |
{
|
| 189 |
+
"epoch": 3.97,
|
|
|
|
| 190 |
"learning_rate": 0.0005,
|
| 191 |
+
"loss": 7.3158,
|
| 192 |
"step": 2100
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 4.0,
|
| 196 |
+
"eval_accuracy": 0.23965490196078432,
|
| 197 |
+
"eval_loss": 7.242833614349365,
|
| 198 |
+
"eval_runtime": 6.6551,
|
| 199 |
+
"eval_samples_per_second": 75.13,
|
| 200 |
+
"eval_steps_per_second": 9.466,
|
| 201 |
"step": 2116
|
| 202 |
},
|
| 203 |
{
|
|
|
|
| 211 |
"step": 2116
|
| 212 |
},
|
| 213 |
{
|
| 214 |
+
"epoch": 4.16,
|
|
|
|
| 215 |
"learning_rate": 0.0005,
|
| 216 |
+
"loss": 7.3307,
|
| 217 |
"step": 2200
|
| 218 |
},
|
| 219 |
{
|
| 220 |
+
"epoch": 4.35,
|
|
|
|
| 221 |
"learning_rate": 0.0005,
|
| 222 |
+
"loss": 7.3307,
|
| 223 |
"step": 2300
|
| 224 |
},
|
| 225 |
{
|
| 226 |
+
"epoch": 4.54,
|
|
|
|
| 227 |
"learning_rate": 0.0005,
|
| 228 |
+
"loss": 7.3197,
|
| 229 |
"step": 2400
|
| 230 |
},
|
| 231 |
{
|
| 232 |
+
"epoch": 4.73,
|
|
|
|
| 233 |
"learning_rate": 0.0005,
|
| 234 |
+
"loss": 7.3238,
|
| 235 |
"step": 2500
|
| 236 |
},
|
| 237 |
{
|
| 238 |
+
"epoch": 4.91,
|
|
|
|
| 239 |
"learning_rate": 0.0005,
|
| 240 |
+
"loss": 7.3017,
|
| 241 |
"step": 2600
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 5.0,
|
| 245 |
+
"eval_accuracy": 0.24474509803921568,
|
| 246 |
+
"eval_loss": 7.229503631591797,
|
| 247 |
+
"eval_runtime": 7.0755,
|
| 248 |
+
"eval_samples_per_second": 70.667,
|
| 249 |
+
"eval_steps_per_second": 8.904,
|
| 250 |
"step": 2645
|
| 251 |
},
|
| 252 |
{
|
|
|
|
| 260 |
"step": 2645
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 5.1,
|
|
|
|
| 264 |
"learning_rate": 0.0005,
|
| 265 |
+
"loss": 7.2926,
|
| 266 |
"step": 2700
|
| 267 |
},
|
| 268 |
{
|
| 269 |
+
"epoch": 5.29,
|
|
|
|
| 270 |
"learning_rate": 0.0005,
|
| 271 |
+
"loss": 7.3006,
|
| 272 |
"step": 2800
|
| 273 |
},
|
| 274 |
{
|
| 275 |
+
"epoch": 5.48,
|
|
|
|
| 276 |
"learning_rate": 0.0005,
|
| 277 |
+
"loss": 7.3051,
|
| 278 |
"step": 2900
|
| 279 |
},
|
| 280 |
{
|
| 281 |
+
"epoch": 5.67,
|
|
|
|
| 282 |
"learning_rate": 0.0005,
|
| 283 |
+
"loss": 7.2973,
|
| 284 |
"step": 3000
|
| 285 |
},
|
| 286 |
{
|
| 287 |
+
"epoch": 5.86,
|
|
|
|
| 288 |
"learning_rate": 0.0005,
|
| 289 |
+
"loss": 7.281,
|
| 290 |
"step": 3100
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 6.0,
|
| 294 |
+
"eval_accuracy": 0.24474509803921568,
|
| 295 |
+
"eval_loss": 7.229503631591797,
|
| 296 |
+
"eval_runtime": 7.1043,
|
| 297 |
+
"eval_samples_per_second": 70.38,
|
| 298 |
+
"eval_steps_per_second": 8.868,
|
| 299 |
"step": 3174
|
| 300 |
},
|
| 301 |
{
|
|
|
|
| 309 |
"step": 3174
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 6.05,
|
|
|
|
| 313 |
"learning_rate": 0.0005,
|
| 314 |
+
"loss": 7.3017,
|
| 315 |
"step": 3200
|
| 316 |
},
|
| 317 |
{
|
| 318 |
+
"epoch": 6.24,
|
|
|
|
| 319 |
"learning_rate": 0.0005,
|
| 320 |
+
"loss": 7.297,
|
| 321 |
"step": 3300
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"epoch": 6.43,
|
|
|
|
| 325 |
"learning_rate": 0.0005,
|
| 326 |
+
"loss": 7.2877,
|
| 327 |
"step": 3400
|
| 328 |
},
|
| 329 |
{
|
| 330 |
+
"epoch": 6.62,
|
|
|
|
| 331 |
"learning_rate": 0.0005,
|
| 332 |
+
"loss": 7.296,
|
| 333 |
"step": 3500
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 6.81,
|
|
|
|
| 337 |
"learning_rate": 0.0005,
|
| 338 |
+
"loss": 7.2993,
|
| 339 |
"step": 3600
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 6.99,
|
|
|
|
| 343 |
"learning_rate": 0.0005,
|
| 344 |
+
"loss": 7.3056,
|
| 345 |
"step": 3700
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 7.0,
|
| 349 |
+
"eval_accuracy": 0.24474509803921568,
|
| 350 |
+
"eval_loss": 7.229503631591797,
|
| 351 |
+
"eval_runtime": 7.6931,
|
| 352 |
+
"eval_samples_per_second": 64.993,
|
| 353 |
+
"eval_steps_per_second": 8.189,
|
| 354 |
"step": 3703
|
| 355 |
},
|
| 356 |
{
|
|
|
|
| 364 |
"step": 3703
|
| 365 |
},
|
| 366 |
{
|
| 367 |
+
"epoch": 7.18,
|
|
|
|
| 368 |
"learning_rate": 0.0005,
|
| 369 |
+
"loss": 7.299,
|
| 370 |
"step": 3800
|
| 371 |
},
|
| 372 |
{
|
| 373 |
+
"epoch": 7.37,
|
|
|
|
| 374 |
"learning_rate": 0.0005,
|
| 375 |
+
"loss": 7.2954,
|
| 376 |
"step": 3900
|
| 377 |
},
|
| 378 |
{
|
| 379 |
+
"epoch": 7.56,
|
|
|
|
| 380 |
"learning_rate": 0.0005,
|
| 381 |
+
"loss": 7.3002,
|
| 382 |
"step": 4000
|
| 383 |
},
|
| 384 |
{
|
| 385 |
+
"epoch": 7.75,
|
|
|
|
| 386 |
"learning_rate": 0.0005,
|
| 387 |
+
"loss": 7.3011,
|
| 388 |
"step": 4100
|
| 389 |
},
|
| 390 |
{
|
| 391 |
+
"epoch": 7.94,
|
|
|
|
| 392 |
"learning_rate": 0.0005,
|
| 393 |
+
"loss": 7.2943,
|
| 394 |
"step": 4200
|
| 395 |
},
|
| 396 |
{
|
| 397 |
"epoch": 8.0,
|
| 398 |
+
"eval_accuracy": 0.24474509803921568,
|
| 399 |
+
"eval_loss": 7.229503631591797,
|
| 400 |
+
"eval_runtime": 6.6465,
|
| 401 |
+
"eval_samples_per_second": 75.228,
|
| 402 |
+
"eval_steps_per_second": 9.479,
|
| 403 |
"step": 4232
|
| 404 |
},
|
| 405 |
{
|
|
|
|
| 413 |
"step": 4232
|
| 414 |
},
|
| 415 |
{
|
| 416 |
+
"epoch": 8.13,
|
|
|
|
| 417 |
"learning_rate": 0.0005,
|
| 418 |
+
"loss": 7.3001,
|
| 419 |
"step": 4300
|
| 420 |
},
|
| 421 |
{
|
| 422 |
+
"epoch": 8.32,
|
|
|
|
| 423 |
"learning_rate": 0.0005,
|
| 424 |
+
"loss": 7.2922,
|
| 425 |
"step": 4400
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"epoch": 8.51,
|
|
|
|
| 429 |
"learning_rate": 0.0005,
|
| 430 |
+
"loss": 7.2969,
|
| 431 |
"step": 4500
|
| 432 |
},
|
| 433 |
{
|
| 434 |
+
"epoch": 8.7,
|
|
|
|
| 435 |
"learning_rate": 0.0005,
|
| 436 |
+
"loss": 7.2879,
|
| 437 |
"step": 4600
|
| 438 |
},
|
| 439 |
{
|
| 440 |
+
"epoch": 8.88,
|
|
|
|
| 441 |
"learning_rate": 0.0005,
|
| 442 |
+
"loss": 7.3011,
|
| 443 |
"step": 4700
|
| 444 |
},
|
| 445 |
{
|
| 446 |
"epoch": 9.0,
|
| 447 |
+
"eval_accuracy": 0.24474509803921568,
|
| 448 |
+
"eval_loss": 7.229503631591797,
|
| 449 |
+
"eval_runtime": 7.77,
|
| 450 |
+
"eval_samples_per_second": 64.35,
|
| 451 |
+
"eval_steps_per_second": 8.108,
|
| 452 |
"step": 4761
|
| 453 |
},
|
| 454 |
{
|
|
|
|
| 462 |
"step": 4761
|
| 463 |
},
|
| 464 |
{
|
| 465 |
+
"epoch": 9.07,
|
|
|
|
| 466 |
"learning_rate": 0.0005,
|
| 467 |
+
"loss": 7.3066,
|
| 468 |
"step": 4800
|
| 469 |
},
|
| 470 |
{
|
| 471 |
+
"epoch": 9.26,
|
|
|
|
| 472 |
"learning_rate": 0.0005,
|
| 473 |
+
"loss": 7.2949,
|
| 474 |
"step": 4900
|
| 475 |
},
|
| 476 |
{
|
| 477 |
+
"epoch": 9.45,
|
|
|
|
| 478 |
"learning_rate": 0.0005,
|
| 479 |
+
"loss": 7.2955,
|
| 480 |
"step": 5000
|
| 481 |
},
|
| 482 |
{
|
| 483 |
+
"epoch": 9.64,
|
|
|
|
| 484 |
"learning_rate": 0.0005,
|
| 485 |
+
"loss": 7.3061,
|
| 486 |
"step": 5100
|
| 487 |
},
|
| 488 |
{
|
| 489 |
+
"epoch": 9.83,
|
|
|
|
| 490 |
"learning_rate": 0.0005,
|
| 491 |
+
"loss": 7.2948,
|
| 492 |
"step": 5200
|
| 493 |
},
|
| 494 |
{
|
| 495 |
"epoch": 10.0,
|
| 496 |
+
"eval_accuracy": 0.24474509803921568,
|
| 497 |
+
"eval_loss": 7.229503631591797,
|
| 498 |
+
"eval_runtime": 7.0663,
|
| 499 |
+
"eval_samples_per_second": 70.758,
|
| 500 |
+
"eval_steps_per_second": 8.916,
|
| 501 |
"step": 5290
|
| 502 |
},
|
| 503 |
{
|
|
|
|
| 511 |
"step": 5290
|
| 512 |
},
|
| 513 |
{
|
| 514 |
+
"epoch": 10.02,
|
|
|
|
| 515 |
"learning_rate": 0.0005,
|
| 516 |
+
"loss": 7.2857,
|
| 517 |
"step": 5300
|
| 518 |
},
|
| 519 |
{
|
| 520 |
+
"epoch": 10.21,
|
|
|
|
| 521 |
"learning_rate": 0.0005,
|
| 522 |
+
"loss": 7.2941,
|
| 523 |
"step": 5400
|
| 524 |
},
|
| 525 |
{
|
| 526 |
+
"epoch": 10.4,
|
|
|
|
| 527 |
"learning_rate": 0.0005,
|
| 528 |
+
"loss": 7.2953,
|
| 529 |
"step": 5500
|
| 530 |
},
|
| 531 |
{
|
| 532 |
+
"epoch": 10.59,
|
|
|
|
| 533 |
"learning_rate": 0.0005,
|
| 534 |
+
"loss": 7.2923,
|
| 535 |
"step": 5600
|
| 536 |
},
|
| 537 |
{
|
| 538 |
+
"epoch": 10.78,
|
|
|
|
| 539 |
"learning_rate": 0.0005,
|
| 540 |
+
"loss": 7.3021,
|
| 541 |
"step": 5700
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 10.96,
|
|
|
|
| 545 |
"learning_rate": 0.0005,
|
| 546 |
+
"loss": 7.3053,
|
| 547 |
"step": 5800
|
| 548 |
},
|
| 549 |
{
|
| 550 |
"epoch": 11.0,
|
| 551 |
+
"eval_accuracy": 0.24474509803921568,
|
| 552 |
+
"eval_loss": 7.229503631591797,
|
| 553 |
+
"eval_runtime": 7.7184,
|
| 554 |
+
"eval_samples_per_second": 64.78,
|
| 555 |
+
"eval_steps_per_second": 8.162,
|
| 556 |
"step": 5819
|
| 557 |
},
|
| 558 |
{
|
|
|
|
| 566 |
"step": 5819
|
| 567 |
},
|
| 568 |
{
|
| 569 |
+
"epoch": 11.15,
|
|
|
|
| 570 |
"learning_rate": 0.0005,
|
| 571 |
+
"loss": 7.2882,
|
| 572 |
"step": 5900
|
| 573 |
},
|
| 574 |
{
|
| 575 |
+
"epoch": 11.34,
|
|
|
|
| 576 |
"learning_rate": 0.0005,
|
| 577 |
+
"loss": 7.3,
|
| 578 |
"step": 6000
|
| 579 |
},
|
| 580 |
{
|
| 581 |
+
"epoch": 11.53,
|
|
|
|
| 582 |
"learning_rate": 0.0005,
|
| 583 |
+
"loss": 7.2984,
|
| 584 |
"step": 6100
|
| 585 |
},
|
| 586 |
{
|
| 587 |
+
"epoch": 11.72,
|
|
|
|
| 588 |
"learning_rate": 0.0005,
|
| 589 |
+
"loss": 7.2966,
|
| 590 |
"step": 6200
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 11.91,
|
|
|
|
| 594 |
"learning_rate": 0.0005,
|
| 595 |
+
"loss": 7.3066,
|
| 596 |
"step": 6300
|
| 597 |
},
|
| 598 |
{
|
| 599 |
"epoch": 12.0,
|
| 600 |
+
"eval_accuracy": 0.24474509803921568,
|
| 601 |
+
"eval_loss": 7.229503631591797,
|
| 602 |
+
"eval_runtime": 7.7125,
|
| 603 |
+
"eval_samples_per_second": 64.83,
|
| 604 |
+
"eval_steps_per_second": 8.169,
|
| 605 |
"step": 6348
|
| 606 |
},
|
| 607 |
{
|
|
|
|
| 615 |
"step": 6348
|
| 616 |
},
|
| 617 |
{
|
| 618 |
+
"epoch": 12.1,
|
|
|
|
| 619 |
"learning_rate": 0.0005,
|
| 620 |
+
"loss": 7.2909,
|
| 621 |
"step": 6400
|
| 622 |
},
|
| 623 |
{
|
| 624 |
+
"epoch": 12.29,
|
|
|
|
| 625 |
"learning_rate": 0.0005,
|
| 626 |
+
"loss": 7.3012,
|
| 627 |
"step": 6500
|
| 628 |
},
|
| 629 |
{
|
| 630 |
+
"epoch": 12.48,
|
|
|
|
| 631 |
"learning_rate": 0.0005,
|
| 632 |
+
"loss": 7.302,
|
| 633 |
"step": 6600
|
| 634 |
},
|
| 635 |
{
|
| 636 |
+
"epoch": 12.67,
|
|
|
|
| 637 |
"learning_rate": 0.0005,
|
| 638 |
+
"loss": 7.2891,
|
| 639 |
"step": 6700
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 12.85,
|
|
|
|
| 643 |
"learning_rate": 0.0005,
|
| 644 |
+
"loss": 7.301,
|
| 645 |
"step": 6800
|
| 646 |
},
|
| 647 |
{
|
| 648 |
"epoch": 13.0,
|
| 649 |
+
"eval_accuracy": 0.24474509803921568,
|
| 650 |
+
"eval_loss": 7.229503631591797,
|
| 651 |
+
"eval_runtime": 7.454,
|
| 652 |
+
"eval_samples_per_second": 67.078,
|
| 653 |
+
"eval_steps_per_second": 8.452,
|
| 654 |
"step": 6877
|
| 655 |
},
|
| 656 |
{
|
|
|
|
| 664 |
"step": 6877
|
| 665 |
},
|
| 666 |
{
|
| 667 |
+
"epoch": 13.04,
|
|
|
|
| 668 |
"learning_rate": 0.0005,
|
| 669 |
+
"loss": 7.2963,
|
| 670 |
"step": 6900
|
| 671 |
},
|
| 672 |
{
|
| 673 |
+
"epoch": 13.23,
|
|
|
|
| 674 |
"learning_rate": 0.0005,
|
| 675 |
+
"loss": 7.292,
|
| 676 |
"step": 7000
|
| 677 |
},
|
| 678 |
{
|
| 679 |
+
"epoch": 13.42,
|
|
|
|
| 680 |
"learning_rate": 0.0005,
|
| 681 |
+
"loss": 7.2949,
|
| 682 |
"step": 7100
|
| 683 |
},
|
| 684 |
{
|
| 685 |
+
"epoch": 13.61,
|
|
|
|
| 686 |
"learning_rate": 0.0005,
|
| 687 |
+
"loss": 7.2904,
|
| 688 |
"step": 7200
|
| 689 |
},
|
| 690 |
{
|
| 691 |
+
"epoch": 13.8,
|
|
|
|
| 692 |
"learning_rate": 0.0005,
|
| 693 |
+
"loss": 7.3018,
|
| 694 |
"step": 7300
|
| 695 |
},
|
| 696 |
{
|
| 697 |
+
"epoch": 13.99,
|
|
|
|
| 698 |
"learning_rate": 0.0005,
|
| 699 |
+
"loss": 7.3009,
|
| 700 |
"step": 7400
|
| 701 |
},
|
| 702 |
{
|
| 703 |
"epoch": 14.0,
|
| 704 |
+
"eval_accuracy": 0.24474509803921568,
|
| 705 |
+
"eval_loss": 7.229503631591797,
|
| 706 |
+
"eval_runtime": 7.6605,
|
| 707 |
+
"eval_samples_per_second": 65.27,
|
| 708 |
+
"eval_steps_per_second": 8.224,
|
| 709 |
"step": 7406
|
| 710 |
},
|
| 711 |
{
|
|
|
|
| 719 |
"step": 7406
|
| 720 |
},
|
| 721 |
{
|
| 722 |
+
"epoch": 14.18,
|
|
|
|
| 723 |
"learning_rate": 0.0005,
|
| 724 |
+
"loss": 7.3057,
|
| 725 |
"step": 7500
|
| 726 |
},
|
| 727 |
{
|
| 728 |
+
"epoch": 14.37,
|
|
|
|
| 729 |
"learning_rate": 0.0005,
|
| 730 |
+
"loss": 7.2973,
|
| 731 |
"step": 7600
|
| 732 |
},
|
| 733 |
{
|
| 734 |
+
"epoch": 14.56,
|
|
|
|
| 735 |
"learning_rate": 0.0005,
|
| 736 |
+
"loss": 7.2915,
|
| 737 |
"step": 7700
|
| 738 |
},
|
| 739 |
{
|
| 740 |
+
"epoch": 14.74,
|
|
|
|
| 741 |
"learning_rate": 0.0005,
|
| 742 |
+
"loss": 7.3049,
|
| 743 |
"step": 7800
|
| 744 |
},
|
| 745 |
{
|
| 746 |
+
"epoch": 14.93,
|
|
|
|
| 747 |
"learning_rate": 0.0005,
|
| 748 |
+
"loss": 7.2906,
|
| 749 |
"step": 7900
|
| 750 |
},
|
| 751 |
{
|
| 752 |
"epoch": 15.0,
|
| 753 |
+
"eval_accuracy": 0.24474509803921568,
|
| 754 |
+
"eval_loss": 7.229503631591797,
|
| 755 |
+
"eval_runtime": 7.6567,
|
| 756 |
+
"eval_samples_per_second": 65.302,
|
| 757 |
+
"eval_steps_per_second": 8.228,
|
| 758 |
"step": 7935
|
| 759 |
},
|
| 760 |
{
|
|
|
|
| 768 |
"step": 7935
|
| 769 |
},
|
| 770 |
{
|
| 771 |
+
"epoch": 15.12,
|
|
|
|
| 772 |
"learning_rate": 0.0005,
|
| 773 |
+
"loss": 7.2845,
|
| 774 |
"step": 8000
|
| 775 |
},
|
| 776 |
{
|
| 777 |
+
"epoch": 15.31,
|
|
|
|
| 778 |
"learning_rate": 0.0005,
|
| 779 |
+
"loss": 7.2956,
|
| 780 |
"step": 8100
|
| 781 |
},
|
| 782 |
{
|
| 783 |
+
"epoch": 15.5,
|
|
|
|
| 784 |
"learning_rate": 0.0005,
|
| 785 |
+
"loss": 7.2983,
|
| 786 |
"step": 8200
|
| 787 |
},
|
| 788 |
{
|
| 789 |
+
"epoch": 15.69,
|
|
|
|
| 790 |
"learning_rate": 0.0005,
|
| 791 |
+
"loss": 7.2957,
|
| 792 |
"step": 8300
|
| 793 |
},
|
| 794 |
{
|
| 795 |
+
"epoch": 15.88,
|
|
|
|
| 796 |
"learning_rate": 0.0005,
|
| 797 |
+
"loss": 7.2955,
|
| 798 |
"step": 8400
|
| 799 |
},
|
| 800 |
{
|
| 801 |
"epoch": 16.0,
|
| 802 |
+
"eval_accuracy": 0.24474509803921568,
|
| 803 |
+
"eval_loss": 7.229503631591797,
|
| 804 |
+
"eval_runtime": 7.5013,
|
| 805 |
+
"eval_samples_per_second": 66.655,
|
| 806 |
+
"eval_steps_per_second": 8.399,
|
| 807 |
"step": 8464
|
| 808 |
},
|
| 809 |
{
|
|
|
|
| 817 |
"step": 8464
|
| 818 |
},
|
| 819 |
{
|
| 820 |
+
"epoch": 16.07,
|
|
|
|
| 821 |
"learning_rate": 0.0005,
|
| 822 |
+
"loss": 7.3081,
|
| 823 |
"step": 8500
|
| 824 |
},
|
| 825 |
{
|
| 826 |
+
"epoch": 16.26,
|
|
|
|
| 827 |
"learning_rate": 0.0005,
|
| 828 |
+
"loss": 7.2922,
|
| 829 |
"step": 8600
|
| 830 |
},
|
| 831 |
{
|
| 832 |
+
"epoch": 16.45,
|
|
|
|
| 833 |
"learning_rate": 0.0005,
|
| 834 |
+
"loss": 7.2993,
|
| 835 |
"step": 8700
|
| 836 |
},
|
| 837 |
{
|
| 838 |
+
"epoch": 16.64,
|
|
|
|
| 839 |
"learning_rate": 0.0005,
|
| 840 |
+
"loss": 7.2944,
|
| 841 |
"step": 8800
|
| 842 |
},
|
| 843 |
{
|
| 844 |
+
"epoch": 16.82,
|
|
|
|
| 845 |
"learning_rate": 0.0005,
|
| 846 |
+
"loss": 7.297,
|
| 847 |
"step": 8900
|
| 848 |
},
|
| 849 |
{
|
| 850 |
"epoch": 17.0,
|
| 851 |
+
"eval_accuracy": 0.24474509803921568,
|
| 852 |
+
"eval_loss": 7.229503631591797,
|
| 853 |
+
"eval_runtime": 6.657,
|
| 854 |
+
"eval_samples_per_second": 75.108,
|
| 855 |
+
"eval_steps_per_second": 9.464,
|
| 856 |
"step": 8993
|
| 857 |
},
|
| 858 |
{
|
|
|
|
| 866 |
"step": 8993
|
| 867 |
},
|
| 868 |
{
|
| 869 |
+
"epoch": 17.01,
|
|
|
|
| 870 |
"learning_rate": 0.0005,
|
| 871 |
+
"loss": 7.3029,
|
| 872 |
"step": 9000
|
| 873 |
},
|
| 874 |
{
|
| 875 |
+
"epoch": 17.2,
|
|
|
|
| 876 |
"learning_rate": 0.0005,
|
| 877 |
+
"loss": 7.3001,
|
| 878 |
"step": 9100
|
| 879 |
},
|
| 880 |
{
|
| 881 |
+
"epoch": 17.39,
|
|
|
|
| 882 |
"learning_rate": 0.0005,
|
| 883 |
+
"loss": 7.3075,
|
| 884 |
"step": 9200
|
| 885 |
},
|
| 886 |
{
|
| 887 |
+
"epoch": 17.58,
|
|
|
|
| 888 |
"learning_rate": 0.0005,
|
| 889 |
+
"loss": 7.2936,
|
| 890 |
"step": 9300
|
| 891 |
},
|
| 892 |
{
|
| 893 |
+
"epoch": 17.77,
|
|
|
|
| 894 |
"learning_rate": 0.0005,
|
| 895 |
+
"loss": 7.2854,
|
| 896 |
"step": 9400
|
| 897 |
},
|
| 898 |
{
|
| 899 |
+
"epoch": 17.96,
|
|
|
|
| 900 |
"learning_rate": 0.0005,
|
| 901 |
+
"loss": 7.2948,
|
| 902 |
"step": 9500
|
| 903 |
},
|
| 904 |
{
|
| 905 |
"epoch": 18.0,
|
| 906 |
+
"eval_accuracy": 0.24474509803921568,
|
| 907 |
+
"eval_loss": 7.229503631591797,
|
| 908 |
+
"eval_runtime": 7.4651,
|
| 909 |
+
"eval_samples_per_second": 66.979,
|
| 910 |
+
"eval_steps_per_second": 8.439,
|
| 911 |
"step": 9522
|
| 912 |
},
|
| 913 |
{
|
|
|
|
| 921 |
"step": 9522
|
| 922 |
},
|
| 923 |
{
|
| 924 |
+
"epoch": 18.15,
|
|
|
|
| 925 |
"learning_rate": 0.0005,
|
| 926 |
+
"loss": 7.309,
|
| 927 |
"step": 9600
|
| 928 |
},
|
| 929 |
{
|
| 930 |
+
"epoch": 18.34,
|
|
|
|
| 931 |
"learning_rate": 0.0005,
|
| 932 |
+
"loss": 7.2911,
|
| 933 |
"step": 9700
|
| 934 |
},
|
| 935 |
{
|
| 936 |
+
"epoch": 18.53,
|
|
|
|
| 937 |
"learning_rate": 0.0005,
|
| 938 |
+
"loss": 7.2851,
|
| 939 |
"step": 9800
|
| 940 |
},
|
| 941 |
{
|
| 942 |
+
"epoch": 18.71,
|
|
|
|
| 943 |
"learning_rate": 0.0005,
|
| 944 |
+
"loss": 7.2975,
|
| 945 |
"step": 9900
|
| 946 |
},
|
| 947 |
{
|
| 948 |
+
"epoch": 18.9,
|
|
|
|
| 949 |
"learning_rate": 0.0005,
|
| 950 |
+
"loss": 7.3039,
|
| 951 |
"step": 10000
|
| 952 |
},
|
| 953 |
{
|
| 954 |
"epoch": 19.0,
|
| 955 |
+
"eval_accuracy": 0.24474509803921568,
|
| 956 |
+
"eval_loss": 7.229503631591797,
|
| 957 |
+
"eval_runtime": 7.6572,
|
| 958 |
+
"eval_samples_per_second": 65.298,
|
| 959 |
+
"eval_steps_per_second": 8.228,
|
| 960 |
"step": 10051
|
| 961 |
},
|
| 962 |
{
|
|
|
|
| 970 |
"step": 10051
|
| 971 |
},
|
| 972 |
{
|
| 973 |
+
"epoch": 19.09,
|
|
|
|
| 974 |
"learning_rate": 0.0005,
|
| 975 |
+
"loss": 7.3074,
|
| 976 |
"step": 10100
|
| 977 |
},
|
| 978 |
{
|
| 979 |
+
"epoch": 19.28,
|
|
|
|
| 980 |
"learning_rate": 0.0005,
|
| 981 |
+
"loss": 7.299,
|
| 982 |
"step": 10200
|
| 983 |
},
|
| 984 |
{
|
| 985 |
+
"epoch": 19.47,
|
|
|
|
| 986 |
"learning_rate": 0.0005,
|
| 987 |
+
"loss": 7.2934,
|
| 988 |
"step": 10300
|
| 989 |
},
|
| 990 |
{
|
| 991 |
+
"epoch": 19.66,
|
|
|
|
| 992 |
"learning_rate": 0.0005,
|
| 993 |
+
"loss": 7.2955,
|
| 994 |
"step": 10400
|
| 995 |
},
|
| 996 |
{
|
| 997 |
+
"epoch": 19.85,
|
|
|
|
| 998 |
"learning_rate": 0.0005,
|
| 999 |
+
"loss": 7.3022,
|
| 1000 |
"step": 10500
|
| 1001 |
},
|
| 1002 |
{
|
| 1003 |
"epoch": 20.0,
|
| 1004 |
+
"eval_accuracy": 0.24474509803921568,
|
| 1005 |
+
"eval_loss": 7.229503631591797,
|
| 1006 |
+
"eval_runtime": 6.6537,
|
| 1007 |
+
"eval_samples_per_second": 75.146,
|
| 1008 |
+
"eval_steps_per_second": 9.468,
|
| 1009 |
"step": 10580
|
| 1010 |
},
|
| 1011 |
{
|
|
|
|
| 1019 |
"step": 10580
|
| 1020 |
},
|
| 1021 |
{
|
| 1022 |
+
"epoch": 20.04,
|
|
|
|
| 1023 |
"learning_rate": 0.0005,
|
| 1024 |
+
"loss": 7.2888,
|
| 1025 |
"step": 10600
|
| 1026 |
},
|
| 1027 |
{
|
| 1028 |
+
"epoch": 20.23,
|
|
|
|
| 1029 |
"learning_rate": 0.0005,
|
| 1030 |
+
"loss": 7.2879,
|
| 1031 |
"step": 10700
|
| 1032 |
},
|
| 1033 |
{
|
| 1034 |
+
"epoch": 20.42,
|
|
|
|
| 1035 |
"learning_rate": 0.0005,
|
| 1036 |
+
"loss": 7.3106,
|
| 1037 |
"step": 10800
|
| 1038 |
},
|
| 1039 |
{
|
| 1040 |
+
"epoch": 20.6,
|
|
|
|
| 1041 |
"learning_rate": 0.0005,
|
| 1042 |
+
"loss": 7.3045,
|
| 1043 |
"step": 10900
|
| 1044 |
},
|
| 1045 |
{
|
| 1046 |
+
"epoch": 20.79,
|
|
|
|
| 1047 |
"learning_rate": 0.0005,
|
| 1048 |
+
"loss": 7.3027,
|
| 1049 |
"step": 11000
|
| 1050 |
},
|
| 1051 |
{
|
| 1052 |
+
"epoch": 20.98,
|
|
|
|
| 1053 |
"learning_rate": 0.0005,
|
| 1054 |
+
"loss": 7.2816,
|
| 1055 |
"step": 11100
|
| 1056 |
},
|
| 1057 |
{
|
| 1058 |
"epoch": 21.0,
|
| 1059 |
+
"eval_accuracy": 0.24474509803921568,
|
| 1060 |
+
"eval_loss": 7.229503631591797,
|
| 1061 |
+
"eval_runtime": 7.4596,
|
| 1062 |
+
"eval_samples_per_second": 67.027,
|
| 1063 |
+
"eval_steps_per_second": 8.445,
|
| 1064 |
"step": 11109
|
| 1065 |
},
|
| 1066 |
{
|
|
|
|
| 1076 |
],
|
| 1077 |
"logging_steps": 100,
|
| 1078 |
"max_steps": 26450,
|
|
|
|
| 1079 |
"num_train_epochs": 50,
|
| 1080 |
"save_steps": 500,
|
| 1081 |
+
"total_flos": 2.3549217117715825e+18,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1082 |
"trial_name": null,
|
| 1083 |
"trial_params": null
|
| 1084 |
}
|
checkpoint-11109/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:b01f751e6b30f6919dad64453b51b9e04d3c36ac30d2f54e574df56cf37b82bc
|
| 3 |
+
size 4728
|