Upload folder using huggingface_hub
Browse files- all_results.json +5 -5
- checkpoint-270/global_step270/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt +1 -1
- checkpoint-270/global_step270/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt +1 -1
- checkpoint-270/global_step270/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt +1 -1
- checkpoint-270/global_step270/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt +1 -1
- checkpoint-270/model-00001-of-00004.safetensors +1 -1
- checkpoint-270/model-00002-of-00004.safetensors +1 -1
- checkpoint-270/model-00003-of-00004.safetensors +1 -1
- checkpoint-270/model-00004-of-00004.safetensors +1 -1
- checkpoint-270/tokenizer.json +2 -2
- checkpoint-270/trainer_state.json +110 -110
- checkpoint-270/training_args.bin +1 -1
- model-00001-of-00004.safetensors +1 -1
- model-00002-of-00004.safetensors +1 -1
- model-00003-of-00004.safetensors +1 -1
- model-00004-of-00004.safetensors +1 -1
- tokenizer.json +2 -2
- train_results.json +5 -5
- trainer_state.json +115 -115
- training_args.bin +1 -1
all_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 0.9975369458128078,
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 6493,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 0.9975369458128078,
|
| 3 |
+
"total_flos": 8.643970128528015e+17,
|
| 4 |
+
"train_loss": 0.07066047384783074,
|
| 5 |
+
"train_runtime": 3995.5402,
|
| 6 |
"train_samples": 6493,
|
| 7 |
+
"train_samples_per_second": 1.625,
|
| 8 |
+
"train_steps_per_second": 0.068
|
| 9 |
}
|
checkpoint-270/global_step270/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 22846855742
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a17f241c3739c198d81e3141056df94bb779dd757a21f9f89af53204d04cfc5e
|
| 3 |
size 22846855742
|
checkpoint-270/global_step270/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 22846855742
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0ee0a96c67870843152695939ea59ac5025f4e3d225b51b3f1a26e8b1f1ac462
|
| 3 |
size 22846855742
|
checkpoint-270/global_step270/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 22846855742
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:21b65f0ec86bbf10a81af977328abd2d86aa4a03f299d85c7db6aa55c0fd14df
|
| 3 |
size 22846855742
|
checkpoint-270/global_step270/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 22846855742
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:20131606e9a33434e3b92633c89bc129e2e60eba86f8236193a952b7a58de305
|
| 3 |
size 22846855742
|
checkpoint-270/model-00001-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4877660776
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2577dc0dbdd5aeedf99725bf6e4df1233bacd820fee699f17a2c80a4b53ce29f
|
| 3 |
size 4877660776
|
checkpoint-270/model-00002-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4932751008
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3d903975881e7178cedcf8c3b6dc3994cae9143f64293260912c927341aa533a
|
| 3 |
size 4932751008
|
checkpoint-270/model-00003-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4330865200
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8a405c24d1e443985effa911704afeb8b8895263c4ca3dd7454eeb3de4cef8b3
|
| 3 |
size 4330865200
|
checkpoint-270/model-00004-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1089994880
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:47537180f9b5498e5481a81d4e10154e932f26f4c2237edbf8d69850e580ed4c
|
| 3 |
size 1089994880
|
checkpoint-270/tokenizer.json
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:bb27d51a5fa5caa8502d091726ff7f63ada64f766ff94afe49fde7d3faba216f
|
| 3 |
+
size 11421996
|
checkpoint-270/trainer_state.json
CHANGED
|
@@ -10,385 +10,385 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.003694581280788177,
|
| 13 |
-
"grad_norm": 1.
|
| 14 |
"learning_rate": 3.7037037037037036e-07,
|
| 15 |
-
"loss": 0.
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.01847290640394089,
|
| 20 |
-
"grad_norm":
|
| 21 |
"learning_rate": 1.8518518518518519e-06,
|
| 22 |
-
"loss": 0.
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.03694581280788178,
|
| 27 |
-
"grad_norm": 0.
|
| 28 |
"learning_rate": 3.7037037037037037e-06,
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.05541871921182266,
|
| 34 |
-
"grad_norm": 0.
|
| 35 |
"learning_rate": 5.555555555555557e-06,
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.07389162561576355,
|
| 41 |
-
"grad_norm": 0.
|
| 42 |
"learning_rate": 7.4074074074074075e-06,
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.09236453201970443,
|
| 48 |
-
"grad_norm": 0.
|
| 49 |
"learning_rate": 9.25925925925926e-06,
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.11083743842364532,
|
| 55 |
-
"grad_norm": 0.
|
| 56 |
"learning_rate": 9.996239762521152e-06,
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.12931034482758622,
|
| 62 |
-
"grad_norm": 0.
|
| 63 |
"learning_rate": 9.973281012033009e-06,
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.1477832512315271,
|
| 69 |
-
"grad_norm": 0.
|
| 70 |
"learning_rate": 9.929548316723983e-06,
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.16625615763546797,
|
| 76 |
-
"grad_norm": 0.
|
| 77 |
"learning_rate": 9.86522435289912e-06,
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.18472906403940886,
|
| 83 |
-
"grad_norm": 0.
|
| 84 |
"learning_rate": 9.7805778088694e-06,
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.20320197044334976,
|
| 90 |
-
"grad_norm": 0.
|
| 91 |
"learning_rate": 9.67596226261095e-06,
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.22167487684729065,
|
| 97 |
-
"grad_norm": 0.
|
| 98 |
"learning_rate": 9.551814704830734e-06,
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.24014778325123154,
|
| 104 |
-
"grad_norm": 0.
|
| 105 |
"learning_rate": 9.40865371360804e-06,
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.25862068965517243,
|
| 111 |
-
"grad_norm": 0.
|
| 112 |
"learning_rate": 9.247077288236488e-06,
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.2770935960591133,
|
| 118 |
-
"grad_norm": 0.
|
| 119 |
"learning_rate": 9.067760351314838e-06,
|
| 120 |
-
"loss": 0.
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.2955665024630542,
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
"learning_rate": 8.871451929520662e-06,
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.31403940886699505,
|
| 132 |
-
"grad_norm": 0.
|
| 133 |
"learning_rate": 8.658972024843063e-06,
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.33251231527093594,
|
| 139 |
-
"grad_norm": 0.
|
| 140 |
"learning_rate": 8.43120818934367e-06,
|
| 141 |
-
"loss": 0.
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.35098522167487683,
|
| 146 |
-
"grad_norm": 0.
|
| 147 |
"learning_rate": 8.18911181775353e-06,
|
| 148 |
-
"loss": 0.
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.3694581280788177,
|
| 153 |
-
"grad_norm": 0.
|
| 154 |
"learning_rate": 7.93369417339209e-06,
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 0.3879310344827586,
|
| 160 |
-
"grad_norm": 0.
|
| 161 |
"learning_rate": 7.666022164008458e-06,
|
| 162 |
-
"loss": 0.
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 0.4064039408866995,
|
| 167 |
-
"grad_norm": 0.
|
| 168 |
"learning_rate": 7.387213885189746e-06,
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.4248768472906404,
|
| 174 |
-
"grad_norm": 0.
|
| 175 |
"learning_rate": 7.098433949952146e-06,
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.4433497536945813,
|
| 181 |
-
"grad_norm": 0.
|
| 182 |
"learning_rate": 6.800888624023552e-06,
|
| 183 |
-
"loss": 0.
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 0.4618226600985222,
|
| 188 |
-
"grad_norm": 0.
|
| 189 |
"learning_rate": 6.495820787138209e-06,
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.4802955665024631,
|
| 195 |
-
"grad_norm": 0.
|
| 196 |
"learning_rate": 6.184504741390596e-06,
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 0.4987684729064039,
|
| 202 |
-
"grad_norm": 0.
|
| 203 |
"learning_rate": 5.8682408883346535e-06,
|
| 204 |
-
"loss": 0.
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.5172413793103449,
|
| 209 |
-
"grad_norm": 0.
|
| 210 |
"learning_rate": 5.548350297062659e-06,
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.5357142857142857,
|
| 216 |
-
"grad_norm": 0.
|
| 217 |
"learning_rate": 5.2261691859535325e-06,
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.5541871921182266,
|
| 223 |
-
"grad_norm": 0.
|
| 224 |
"learning_rate": 4.903043341140879e-06,
|
| 225 |
-
"loss": 0.
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 0.5726600985221675,
|
| 230 |
-
"grad_norm": 0.
|
| 231 |
"learning_rate": 4.580322495015466e-06,
|
| 232 |
-
"loss": 0.
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.5911330049261084,
|
| 237 |
-
"grad_norm": 0.
|
| 238 |
"learning_rate": 4.259354688243758e-06,
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 0.6096059113300493,
|
| 244 |
-
"grad_norm": 0.
|
| 245 |
"learning_rate": 3.941480638852948e-06,
|
| 246 |
-
"loss": 0.
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"epoch": 0.6280788177339901,
|
| 251 |
-
"grad_norm": 0.
|
| 252 |
"learning_rate": 3.6280281419034934e-06,
|
| 253 |
-
"loss": 0.
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 0.646551724137931,
|
| 258 |
-
"grad_norm": 0.
|
| 259 |
"learning_rate": 3.3203065231422904e-06,
|
| 260 |
-
"loss": 0.
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 0.6650246305418719,
|
| 265 |
-
"grad_norm": 0.
|
| 266 |
"learning_rate": 3.019601169804216e-06,
|
| 267 |
-
"loss": 0.
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
"epoch": 0.6834975369458128,
|
| 272 |
-
"grad_norm": 0.
|
| 273 |
"learning_rate": 2.7271681614074973e-06,
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 0.7019704433497537,
|
| 279 |
-
"grad_norm": 0.
|
| 280 |
"learning_rate": 2.4442290229706344e-06,
|
| 281 |
-
"loss": 0.
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 0.7204433497536946,
|
| 286 |
-
"grad_norm": 0.
|
| 287 |
"learning_rate": 2.171965622567308e-06,
|
| 288 |
-
"loss": 0.
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.7389162561576355,
|
| 293 |
-
"grad_norm": 0.
|
| 294 |
"learning_rate": 1.9115152345327154e-06,
|
| 295 |
-
"loss": 0.
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
"epoch": 0.7573891625615764,
|
| 300 |
-
"grad_norm": 0.
|
| 301 |
"learning_rate": 1.6639657889429017e-06,
|
| 302 |
-
"loss": 0.
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 0.7758620689655172,
|
| 307 |
-
"grad_norm": 0.
|
| 308 |
"learning_rate": 1.4303513272105057e-06,
|
| 309 |
-
"loss": 0.
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 0.7943349753694581,
|
| 314 |
-
"grad_norm": 0.
|
| 315 |
"learning_rate": 1.2116476827794104e-06,
|
| 316 |
-
"loss": 0.
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 0.812807881773399,
|
| 321 |
-
"grad_norm": 0.
|
| 322 |
"learning_rate": 1.008768404960535e-06,
|
| 323 |
-
"loss": 0.
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 0.8312807881773399,
|
| 328 |
-
"grad_norm": 0.
|
| 329 |
"learning_rate": 8.225609429353187e-07,
|
| 330 |
-
"loss": 0.
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 0.8497536945812808,
|
| 335 |
-
"grad_norm": 0.
|
| 336 |
"learning_rate": 6.53803105866761e-07,
|
| 337 |
-
"loss": 0.
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 0.8682266009852216,
|
| 342 |
-
"grad_norm": 0.
|
| 343 |
"learning_rate": 5.031998139045352e-07,
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.8866995073891626,
|
| 349 |
-
"grad_norm": 0.
|
| 350 |
"learning_rate": 3.7138015365554834e-07,
|
| 351 |
-
"loss": 0.
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
"epoch": 0.9051724137931034,
|
| 356 |
-
"grad_norm": 0.
|
| 357 |
"learning_rate": 2.5889475041961767e-07,
|
| 358 |
-
"loss": 0.
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 0.9236453201970444,
|
| 363 |
-
"grad_norm": 0.
|
| 364 |
"learning_rate": 1.6621346816668993e-07,
|
| 365 |
-
"loss": 0.
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
"epoch": 0.9421182266009852,
|
| 370 |
-
"grad_norm": 0.
|
| 371 |
"learning_rate": 9.372344686307655e-08,
|
| 372 |
-
"loss": 0.
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
"epoch": 0.9605911330049262,
|
| 377 |
-
"grad_norm": 0.
|
| 378 |
"learning_rate": 4.172748534499449e-08,
|
| 379 |
-
"loss": 0.
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
"epoch": 0.979064039408867,
|
| 384 |
-
"grad_norm": 0.
|
| 385 |
"learning_rate": 1.044277649433989e-08,
|
| 386 |
-
"loss": 0.
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
"epoch": 0.9975369458128078,
|
| 391 |
-
"grad_norm": 0.
|
| 392 |
"learning_rate": 0.0,
|
| 393 |
"loss": 0.033,
|
| 394 |
"step": 270
|
|
@@ -411,7 +411,7 @@
|
|
| 411 |
"attributes": {}
|
| 412 |
}
|
| 413 |
},
|
| 414 |
-
"total_flos":
|
| 415 |
"train_batch_size": 1,
|
| 416 |
"trial_name": null,
|
| 417 |
"trial_params": null
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.003694581280788177,
|
| 13 |
+
"grad_norm": 1.3164679266754151,
|
| 14 |
"learning_rate": 3.7037037037037036e-07,
|
| 15 |
+
"loss": 0.5867,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.01847290640394089,
|
| 20 |
+
"grad_norm": 0.8941813303905811,
|
| 21 |
"learning_rate": 1.8518518518518519e-06,
|
| 22 |
+
"loss": 0.5684,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.03694581280788178,
|
| 27 |
+
"grad_norm": 0.4848149582760561,
|
| 28 |
"learning_rate": 3.7037037037037037e-06,
|
| 29 |
+
"loss": 0.5272,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.05541871921182266,
|
| 34 |
+
"grad_norm": 0.23489374203812283,
|
| 35 |
"learning_rate": 5.555555555555557e-06,
|
| 36 |
+
"loss": 0.3213,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.07389162561576355,
|
| 41 |
+
"grad_norm": 0.17590513475537736,
|
| 42 |
"learning_rate": 7.4074074074074075e-06,
|
| 43 |
+
"loss": 0.1719,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.09236453201970443,
|
| 48 |
+
"grad_norm": 0.09924337039651394,
|
| 49 |
"learning_rate": 9.25925925925926e-06,
|
| 50 |
+
"loss": 0.1392,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.11083743842364532,
|
| 55 |
+
"grad_norm": 0.06829427058420333,
|
| 56 |
"learning_rate": 9.996239762521152e-06,
|
| 57 |
+
"loss": 0.1168,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.12931034482758622,
|
| 62 |
+
"grad_norm": 0.06324690561682543,
|
| 63 |
"learning_rate": 9.973281012033009e-06,
|
| 64 |
+
"loss": 0.1009,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.1477832512315271,
|
| 69 |
+
"grad_norm": 0.049072943668117305,
|
| 70 |
"learning_rate": 9.929548316723983e-06,
|
| 71 |
+
"loss": 0.0886,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.16625615763546797,
|
| 76 |
+
"grad_norm": 0.04472692512923622,
|
| 77 |
"learning_rate": 9.86522435289912e-06,
|
| 78 |
+
"loss": 0.0772,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.18472906403940886,
|
| 83 |
+
"grad_norm": 0.04320806158867543,
|
| 84 |
"learning_rate": 9.7805778088694e-06,
|
| 85 |
+
"loss": 0.0677,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.20320197044334976,
|
| 90 |
+
"grad_norm": 0.04347088345515666,
|
| 91 |
"learning_rate": 9.67596226261095e-06,
|
| 92 |
+
"loss": 0.0652,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.22167487684729065,
|
| 97 |
+
"grad_norm": 0.04636829853932609,
|
| 98 |
"learning_rate": 9.551814704830734e-06,
|
| 99 |
+
"loss": 0.0618,
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.24014778325123154,
|
| 104 |
+
"grad_norm": 0.04911651390500423,
|
| 105 |
"learning_rate": 9.40865371360804e-06,
|
| 106 |
+
"loss": 0.0564,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.25862068965517243,
|
| 111 |
+
"grad_norm": 0.046124586761473525,
|
| 112 |
"learning_rate": 9.247077288236488e-06,
|
| 113 |
+
"loss": 0.0496,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.2770935960591133,
|
| 118 |
+
"grad_norm": 0.04282690739591252,
|
| 119 |
"learning_rate": 9.067760351314838e-06,
|
| 120 |
+
"loss": 0.0443,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.2955665024630542,
|
| 125 |
+
"grad_norm": 0.0501495787415551,
|
| 126 |
"learning_rate": 8.871451929520662e-06,
|
| 127 |
+
"loss": 0.0472,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.31403940886699505,
|
| 132 |
+
"grad_norm": 0.048856614296215864,
|
| 133 |
"learning_rate": 8.658972024843063e-06,
|
| 134 |
+
"loss": 0.045,
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.33251231527093594,
|
| 139 |
+
"grad_norm": 0.052135344528722635,
|
| 140 |
"learning_rate": 8.43120818934367e-06,
|
| 141 |
+
"loss": 0.0407,
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.35098522167487683,
|
| 146 |
+
"grad_norm": 0.04496874450828456,
|
| 147 |
"learning_rate": 8.18911181775353e-06,
|
| 148 |
+
"loss": 0.0443,
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.3694581280788177,
|
| 153 |
+
"grad_norm": 0.046961464584805046,
|
| 154 |
"learning_rate": 7.93369417339209e-06,
|
| 155 |
+
"loss": 0.043,
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 0.3879310344827586,
|
| 160 |
+
"grad_norm": 0.03774079876533218,
|
| 161 |
"learning_rate": 7.666022164008458e-06,
|
| 162 |
+
"loss": 0.039,
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 0.4064039408866995,
|
| 167 |
+
"grad_norm": 0.039388091387549375,
|
| 168 |
"learning_rate": 7.387213885189746e-06,
|
| 169 |
+
"loss": 0.043,
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.4248768472906404,
|
| 174 |
+
"grad_norm": 0.04010524552891231,
|
| 175 |
"learning_rate": 7.098433949952146e-06,
|
| 176 |
+
"loss": 0.0418,
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.4433497536945813,
|
| 181 |
+
"grad_norm": 0.03364245597783716,
|
| 182 |
"learning_rate": 6.800888624023552e-06,
|
| 183 |
+
"loss": 0.0396,
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 0.4618226600985222,
|
| 188 |
+
"grad_norm": 0.03214650791918716,
|
| 189 |
"learning_rate": 6.495820787138209e-06,
|
| 190 |
+
"loss": 0.0343,
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.4802955665024631,
|
| 195 |
+
"grad_norm": 0.04248682387196562,
|
| 196 |
"learning_rate": 6.184504741390596e-06,
|
| 197 |
+
"loss": 0.0385,
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 0.4987684729064039,
|
| 202 |
+
"grad_norm": 0.041745690695414894,
|
| 203 |
"learning_rate": 5.8682408883346535e-06,
|
| 204 |
+
"loss": 0.0354,
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.5172413793103449,
|
| 209 |
+
"grad_norm": 0.043998119847084,
|
| 210 |
"learning_rate": 5.548350297062659e-06,
|
| 211 |
+
"loss": 0.0363,
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.5357142857142857,
|
| 216 |
+
"grad_norm": 0.03523911995673237,
|
| 217 |
"learning_rate": 5.2261691859535325e-06,
|
| 218 |
+
"loss": 0.0311,
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.5541871921182266,
|
| 223 |
+
"grad_norm": 0.051683301339415226,
|
| 224 |
"learning_rate": 4.903043341140879e-06,
|
| 225 |
+
"loss": 0.0322,
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 0.5726600985221675,
|
| 230 |
+
"grad_norm": 0.029645536386539162,
|
| 231 |
"learning_rate": 4.580322495015466e-06,
|
| 232 |
+
"loss": 0.0297,
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.5911330049261084,
|
| 237 |
+
"grad_norm": 0.038478089898929216,
|
| 238 |
"learning_rate": 4.259354688243758e-06,
|
| 239 |
+
"loss": 0.0373,
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 0.6096059113300493,
|
| 244 |
+
"grad_norm": 0.04016397060959619,
|
| 245 |
"learning_rate": 3.941480638852948e-06,
|
| 246 |
+
"loss": 0.0293,
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"epoch": 0.6280788177339901,
|
| 251 |
+
"grad_norm": 0.03295300026030036,
|
| 252 |
"learning_rate": 3.6280281419034934e-06,
|
| 253 |
+
"loss": 0.0306,
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 0.646551724137931,
|
| 258 |
+
"grad_norm": 0.04371251122688708,
|
| 259 |
"learning_rate": 3.3203065231422904e-06,
|
| 260 |
+
"loss": 0.0292,
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 0.6650246305418719,
|
| 265 |
+
"grad_norm": 0.02878432244430226,
|
| 266 |
"learning_rate": 3.019601169804216e-06,
|
| 267 |
+
"loss": 0.0348,
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
"epoch": 0.6834975369458128,
|
| 272 |
+
"grad_norm": 0.03582599925982462,
|
| 273 |
"learning_rate": 2.7271681614074973e-06,
|
| 274 |
+
"loss": 0.0292,
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 0.7019704433497537,
|
| 279 |
+
"grad_norm": 0.04498392884678493,
|
| 280 |
"learning_rate": 2.4442290229706344e-06,
|
| 281 |
+
"loss": 0.0355,
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 0.7204433497536946,
|
| 286 |
+
"grad_norm": 0.03888587884987569,
|
| 287 |
"learning_rate": 2.171965622567308e-06,
|
| 288 |
+
"loss": 0.0311,
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.7389162561576355,
|
| 293 |
+
"grad_norm": 0.04684421314626146,
|
| 294 |
"learning_rate": 1.9115152345327154e-06,
|
| 295 |
+
"loss": 0.0391,
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
"epoch": 0.7573891625615764,
|
| 300 |
+
"grad_norm": 0.031274018167510506,
|
| 301 |
"learning_rate": 1.6639657889429017e-06,
|
| 302 |
+
"loss": 0.0275,
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 0.7758620689655172,
|
| 307 |
+
"grad_norm": 0.03793766745215515,
|
| 308 |
"learning_rate": 1.4303513272105057e-06,
|
| 309 |
+
"loss": 0.0312,
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 0.7943349753694581,
|
| 314 |
+
"grad_norm": 0.03084526133895099,
|
| 315 |
"learning_rate": 1.2116476827794104e-06,
|
| 316 |
+
"loss": 0.0334,
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 0.812807881773399,
|
| 321 |
+
"grad_norm": 0.035909978135080484,
|
| 322 |
"learning_rate": 1.008768404960535e-06,
|
| 323 |
+
"loss": 0.0329,
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 0.8312807881773399,
|
| 328 |
+
"grad_norm": 0.03337500963867465,
|
| 329 |
"learning_rate": 8.225609429353187e-07,
|
| 330 |
+
"loss": 0.0299,
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 0.8497536945812808,
|
| 335 |
+
"grad_norm": 0.04194808091582502,
|
| 336 |
"learning_rate": 6.53803105866761e-07,
|
| 337 |
+
"loss": 0.0312,
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 0.8682266009852216,
|
| 342 |
+
"grad_norm": 0.030185202275612954,
|
| 343 |
"learning_rate": 5.031998139045352e-07,
|
| 344 |
+
"loss": 0.0291,
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.8866995073891626,
|
| 349 |
+
"grad_norm": 0.03678202209697851,
|
| 350 |
"learning_rate": 3.7138015365554834e-07,
|
| 351 |
+
"loss": 0.0329,
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
"epoch": 0.9051724137931034,
|
| 356 |
+
"grad_norm": 0.026908956915127933,
|
| 357 |
"learning_rate": 2.5889475041961767e-07,
|
| 358 |
+
"loss": 0.0318,
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 0.9236453201970444,
|
| 363 |
+
"grad_norm": 0.04036220391547687,
|
| 364 |
"learning_rate": 1.6621346816668993e-07,
|
| 365 |
+
"loss": 0.0304,
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
"epoch": 0.9421182266009852,
|
| 370 |
+
"grad_norm": 0.037246305929994866,
|
| 371 |
"learning_rate": 9.372344686307655e-08,
|
| 372 |
+
"loss": 0.036,
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
"epoch": 0.9605911330049262,
|
| 377 |
+
"grad_norm": 0.031641745281581694,
|
| 378 |
"learning_rate": 4.172748534499449e-08,
|
| 379 |
+
"loss": 0.0296,
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
"epoch": 0.979064039408867,
|
| 384 |
+
"grad_norm": 0.03135655037281048,
|
| 385 |
"learning_rate": 1.044277649433989e-08,
|
| 386 |
+
"loss": 0.0294,
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
"epoch": 0.9975369458128078,
|
| 391 |
+
"grad_norm": 0.03097346411772941,
|
| 392 |
"learning_rate": 0.0,
|
| 393 |
"loss": 0.033,
|
| 394 |
"step": 270
|
|
|
|
| 411 |
"attributes": {}
|
| 412 |
}
|
| 413 |
},
|
| 414 |
+
"total_flos": 8.643970128528015e+17,
|
| 415 |
"train_batch_size": 1,
|
| 416 |
"trial_name": null,
|
| 417 |
"trial_params": null
|
checkpoint-270/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 7352
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68d8b75a1101f455c3bf040fa2f84956565ddbcbbaa9e7e94c618c261996c857
|
| 3 |
size 7352
|
model-00001-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4877660776
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2577dc0dbdd5aeedf99725bf6e4df1233bacd820fee699f17a2c80a4b53ce29f
|
| 3 |
size 4877660776
|
model-00002-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4932751008
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3d903975881e7178cedcf8c3b6dc3994cae9143f64293260912c927341aa533a
|
| 3 |
size 4932751008
|
model-00003-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4330865200
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8a405c24d1e443985effa911704afeb8b8895263c4ca3dd7454eeb3de4cef8b3
|
| 3 |
size 4330865200
|
model-00004-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1089994880
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:47537180f9b5498e5481a81d4e10154e932f26f4c2237edbf8d69850e580ed4c
|
| 3 |
size 1089994880
|
tokenizer.json
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:bb27d51a5fa5caa8502d091726ff7f63ada64f766ff94afe49fde7d3faba216f
|
| 3 |
+
size 11421996
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 0.9975369458128078,
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 6493,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 0.9975369458128078,
|
| 3 |
+
"total_flos": 8.643970128528015e+17,
|
| 4 |
+
"train_loss": 0.07066047384783074,
|
| 5 |
+
"train_runtime": 3995.5402,
|
| 6 |
"train_samples": 6493,
|
| 7 |
+
"train_samples_per_second": 1.625,
|
| 8 |
+
"train_steps_per_second": 0.068
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -10,385 +10,385 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.003694581280788177,
|
| 13 |
-
"grad_norm": 1.
|
| 14 |
"learning_rate": 3.7037037037037036e-07,
|
| 15 |
-
"loss": 0.
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.01847290640394089,
|
| 20 |
-
"grad_norm":
|
| 21 |
"learning_rate": 1.8518518518518519e-06,
|
| 22 |
-
"loss": 0.
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.03694581280788178,
|
| 27 |
-
"grad_norm": 0.
|
| 28 |
"learning_rate": 3.7037037037037037e-06,
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.05541871921182266,
|
| 34 |
-
"grad_norm": 0.
|
| 35 |
"learning_rate": 5.555555555555557e-06,
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.07389162561576355,
|
| 41 |
-
"grad_norm": 0.
|
| 42 |
"learning_rate": 7.4074074074074075e-06,
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.09236453201970443,
|
| 48 |
-
"grad_norm": 0.
|
| 49 |
"learning_rate": 9.25925925925926e-06,
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.11083743842364532,
|
| 55 |
-
"grad_norm": 0.
|
| 56 |
"learning_rate": 9.996239762521152e-06,
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.12931034482758622,
|
| 62 |
-
"grad_norm": 0.
|
| 63 |
"learning_rate": 9.973281012033009e-06,
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.1477832512315271,
|
| 69 |
-
"grad_norm": 0.
|
| 70 |
"learning_rate": 9.929548316723983e-06,
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.16625615763546797,
|
| 76 |
-
"grad_norm": 0.
|
| 77 |
"learning_rate": 9.86522435289912e-06,
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.18472906403940886,
|
| 83 |
-
"grad_norm": 0.
|
| 84 |
"learning_rate": 9.7805778088694e-06,
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.20320197044334976,
|
| 90 |
-
"grad_norm": 0.
|
| 91 |
"learning_rate": 9.67596226261095e-06,
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.22167487684729065,
|
| 97 |
-
"grad_norm": 0.
|
| 98 |
"learning_rate": 9.551814704830734e-06,
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.24014778325123154,
|
| 104 |
-
"grad_norm": 0.
|
| 105 |
"learning_rate": 9.40865371360804e-06,
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.25862068965517243,
|
| 111 |
-
"grad_norm": 0.
|
| 112 |
"learning_rate": 9.247077288236488e-06,
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.2770935960591133,
|
| 118 |
-
"grad_norm": 0.
|
| 119 |
"learning_rate": 9.067760351314838e-06,
|
| 120 |
-
"loss": 0.
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.2955665024630542,
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
"learning_rate": 8.871451929520662e-06,
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.31403940886699505,
|
| 132 |
-
"grad_norm": 0.
|
| 133 |
"learning_rate": 8.658972024843063e-06,
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.33251231527093594,
|
| 139 |
-
"grad_norm": 0.
|
| 140 |
"learning_rate": 8.43120818934367e-06,
|
| 141 |
-
"loss": 0.
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.35098522167487683,
|
| 146 |
-
"grad_norm": 0.
|
| 147 |
"learning_rate": 8.18911181775353e-06,
|
| 148 |
-
"loss": 0.
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.3694581280788177,
|
| 153 |
-
"grad_norm": 0.
|
| 154 |
"learning_rate": 7.93369417339209e-06,
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 0.3879310344827586,
|
| 160 |
-
"grad_norm": 0.
|
| 161 |
"learning_rate": 7.666022164008458e-06,
|
| 162 |
-
"loss": 0.
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 0.4064039408866995,
|
| 167 |
-
"grad_norm": 0.
|
| 168 |
"learning_rate": 7.387213885189746e-06,
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.4248768472906404,
|
| 174 |
-
"grad_norm": 0.
|
| 175 |
"learning_rate": 7.098433949952146e-06,
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.4433497536945813,
|
| 181 |
-
"grad_norm": 0.
|
| 182 |
"learning_rate": 6.800888624023552e-06,
|
| 183 |
-
"loss": 0.
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 0.4618226600985222,
|
| 188 |
-
"grad_norm": 0.
|
| 189 |
"learning_rate": 6.495820787138209e-06,
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.4802955665024631,
|
| 195 |
-
"grad_norm": 0.
|
| 196 |
"learning_rate": 6.184504741390596e-06,
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 0.4987684729064039,
|
| 202 |
-
"grad_norm": 0.
|
| 203 |
"learning_rate": 5.8682408883346535e-06,
|
| 204 |
-
"loss": 0.
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.5172413793103449,
|
| 209 |
-
"grad_norm": 0.
|
| 210 |
"learning_rate": 5.548350297062659e-06,
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.5357142857142857,
|
| 216 |
-
"grad_norm": 0.
|
| 217 |
"learning_rate": 5.2261691859535325e-06,
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.5541871921182266,
|
| 223 |
-
"grad_norm": 0.
|
| 224 |
"learning_rate": 4.903043341140879e-06,
|
| 225 |
-
"loss": 0.
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 0.5726600985221675,
|
| 230 |
-
"grad_norm": 0.
|
| 231 |
"learning_rate": 4.580322495015466e-06,
|
| 232 |
-
"loss": 0.
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.5911330049261084,
|
| 237 |
-
"grad_norm": 0.
|
| 238 |
"learning_rate": 4.259354688243758e-06,
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 0.6096059113300493,
|
| 244 |
-
"grad_norm": 0.
|
| 245 |
"learning_rate": 3.941480638852948e-06,
|
| 246 |
-
"loss": 0.
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"epoch": 0.6280788177339901,
|
| 251 |
-
"grad_norm": 0.
|
| 252 |
"learning_rate": 3.6280281419034934e-06,
|
| 253 |
-
"loss": 0.
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 0.646551724137931,
|
| 258 |
-
"grad_norm": 0.
|
| 259 |
"learning_rate": 3.3203065231422904e-06,
|
| 260 |
-
"loss": 0.
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 0.6650246305418719,
|
| 265 |
-
"grad_norm": 0.
|
| 266 |
"learning_rate": 3.019601169804216e-06,
|
| 267 |
-
"loss": 0.
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
"epoch": 0.6834975369458128,
|
| 272 |
-
"grad_norm": 0.
|
| 273 |
"learning_rate": 2.7271681614074973e-06,
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 0.7019704433497537,
|
| 279 |
-
"grad_norm": 0.
|
| 280 |
"learning_rate": 2.4442290229706344e-06,
|
| 281 |
-
"loss": 0.
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 0.7204433497536946,
|
| 286 |
-
"grad_norm": 0.
|
| 287 |
"learning_rate": 2.171965622567308e-06,
|
| 288 |
-
"loss": 0.
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.7389162561576355,
|
| 293 |
-
"grad_norm": 0.
|
| 294 |
"learning_rate": 1.9115152345327154e-06,
|
| 295 |
-
"loss": 0.
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
"epoch": 0.7573891625615764,
|
| 300 |
-
"grad_norm": 0.
|
| 301 |
"learning_rate": 1.6639657889429017e-06,
|
| 302 |
-
"loss": 0.
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 0.7758620689655172,
|
| 307 |
-
"grad_norm": 0.
|
| 308 |
"learning_rate": 1.4303513272105057e-06,
|
| 309 |
-
"loss": 0.
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 0.7943349753694581,
|
| 314 |
-
"grad_norm": 0.
|
| 315 |
"learning_rate": 1.2116476827794104e-06,
|
| 316 |
-
"loss": 0.
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 0.812807881773399,
|
| 321 |
-
"grad_norm": 0.
|
| 322 |
"learning_rate": 1.008768404960535e-06,
|
| 323 |
-
"loss": 0.
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 0.8312807881773399,
|
| 328 |
-
"grad_norm": 0.
|
| 329 |
"learning_rate": 8.225609429353187e-07,
|
| 330 |
-
"loss": 0.
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 0.8497536945812808,
|
| 335 |
-
"grad_norm": 0.
|
| 336 |
"learning_rate": 6.53803105866761e-07,
|
| 337 |
-
"loss": 0.
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 0.8682266009852216,
|
| 342 |
-
"grad_norm": 0.
|
| 343 |
"learning_rate": 5.031998139045352e-07,
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.8866995073891626,
|
| 349 |
-
"grad_norm": 0.
|
| 350 |
"learning_rate": 3.7138015365554834e-07,
|
| 351 |
-
"loss": 0.
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
"epoch": 0.9051724137931034,
|
| 356 |
-
"grad_norm": 0.
|
| 357 |
"learning_rate": 2.5889475041961767e-07,
|
| 358 |
-
"loss": 0.
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 0.9236453201970444,
|
| 363 |
-
"grad_norm": 0.
|
| 364 |
"learning_rate": 1.6621346816668993e-07,
|
| 365 |
-
"loss": 0.
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
"epoch": 0.9421182266009852,
|
| 370 |
-
"grad_norm": 0.
|
| 371 |
"learning_rate": 9.372344686307655e-08,
|
| 372 |
-
"loss": 0.
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
"epoch": 0.9605911330049262,
|
| 377 |
-
"grad_norm": 0.
|
| 378 |
"learning_rate": 4.172748534499449e-08,
|
| 379 |
-
"loss": 0.
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
"epoch": 0.979064039408867,
|
| 384 |
-
"grad_norm": 0.
|
| 385 |
"learning_rate": 1.044277649433989e-08,
|
| 386 |
-
"loss": 0.
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
"epoch": 0.9975369458128078,
|
| 391 |
-
"grad_norm": 0.
|
| 392 |
"learning_rate": 0.0,
|
| 393 |
"loss": 0.033,
|
| 394 |
"step": 270
|
|
@@ -396,11 +396,11 @@
|
|
| 396 |
{
|
| 397 |
"epoch": 0.9975369458128078,
|
| 398 |
"step": 270,
|
| 399 |
-
"total_flos":
|
| 400 |
-
"train_loss": 0.
|
| 401 |
-
"train_runtime":
|
| 402 |
-
"train_samples_per_second":
|
| 403 |
-
"train_steps_per_second": 0.
|
| 404 |
}
|
| 405 |
],
|
| 406 |
"logging_steps": 5,
|
|
@@ -420,7 +420,7 @@
|
|
| 420 |
"attributes": {}
|
| 421 |
}
|
| 422 |
},
|
| 423 |
-
"total_flos":
|
| 424 |
"train_batch_size": 1,
|
| 425 |
"trial_name": null,
|
| 426 |
"trial_params": null
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.003694581280788177,
|
| 13 |
+
"grad_norm": 1.3164679266754151,
|
| 14 |
"learning_rate": 3.7037037037037036e-07,
|
| 15 |
+
"loss": 0.5867,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.01847290640394089,
|
| 20 |
+
"grad_norm": 0.8941813303905811,
|
| 21 |
"learning_rate": 1.8518518518518519e-06,
|
| 22 |
+
"loss": 0.5684,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.03694581280788178,
|
| 27 |
+
"grad_norm": 0.4848149582760561,
|
| 28 |
"learning_rate": 3.7037037037037037e-06,
|
| 29 |
+
"loss": 0.5272,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.05541871921182266,
|
| 34 |
+
"grad_norm": 0.23489374203812283,
|
| 35 |
"learning_rate": 5.555555555555557e-06,
|
| 36 |
+
"loss": 0.3213,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.07389162561576355,
|
| 41 |
+
"grad_norm": 0.17590513475537736,
|
| 42 |
"learning_rate": 7.4074074074074075e-06,
|
| 43 |
+
"loss": 0.1719,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.09236453201970443,
|
| 48 |
+
"grad_norm": 0.09924337039651394,
|
| 49 |
"learning_rate": 9.25925925925926e-06,
|
| 50 |
+
"loss": 0.1392,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.11083743842364532,
|
| 55 |
+
"grad_norm": 0.06829427058420333,
|
| 56 |
"learning_rate": 9.996239762521152e-06,
|
| 57 |
+
"loss": 0.1168,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.12931034482758622,
|
| 62 |
+
"grad_norm": 0.06324690561682543,
|
| 63 |
"learning_rate": 9.973281012033009e-06,
|
| 64 |
+
"loss": 0.1009,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.1477832512315271,
|
| 69 |
+
"grad_norm": 0.049072943668117305,
|
| 70 |
"learning_rate": 9.929548316723983e-06,
|
| 71 |
+
"loss": 0.0886,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.16625615763546797,
|
| 76 |
+
"grad_norm": 0.04472692512923622,
|
| 77 |
"learning_rate": 9.86522435289912e-06,
|
| 78 |
+
"loss": 0.0772,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.18472906403940886,
|
| 83 |
+
"grad_norm": 0.04320806158867543,
|
| 84 |
"learning_rate": 9.7805778088694e-06,
|
| 85 |
+
"loss": 0.0677,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.20320197044334976,
|
| 90 |
+
"grad_norm": 0.04347088345515666,
|
| 91 |
"learning_rate": 9.67596226261095e-06,
|
| 92 |
+
"loss": 0.0652,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.22167487684729065,
|
| 97 |
+
"grad_norm": 0.04636829853932609,
|
| 98 |
"learning_rate": 9.551814704830734e-06,
|
| 99 |
+
"loss": 0.0618,
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.24014778325123154,
|
| 104 |
+
"grad_norm": 0.04911651390500423,
|
| 105 |
"learning_rate": 9.40865371360804e-06,
|
| 106 |
+
"loss": 0.0564,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.25862068965517243,
|
| 111 |
+
"grad_norm": 0.046124586761473525,
|
| 112 |
"learning_rate": 9.247077288236488e-06,
|
| 113 |
+
"loss": 0.0496,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.2770935960591133,
|
| 118 |
+
"grad_norm": 0.04282690739591252,
|
| 119 |
"learning_rate": 9.067760351314838e-06,
|
| 120 |
+
"loss": 0.0443,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.2955665024630542,
|
| 125 |
+
"grad_norm": 0.0501495787415551,
|
| 126 |
"learning_rate": 8.871451929520662e-06,
|
| 127 |
+
"loss": 0.0472,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.31403940886699505,
|
| 132 |
+
"grad_norm": 0.048856614296215864,
|
| 133 |
"learning_rate": 8.658972024843063e-06,
|
| 134 |
+
"loss": 0.045,
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.33251231527093594,
|
| 139 |
+
"grad_norm": 0.052135344528722635,
|
| 140 |
"learning_rate": 8.43120818934367e-06,
|
| 141 |
+
"loss": 0.0407,
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.35098522167487683,
|
| 146 |
+
"grad_norm": 0.04496874450828456,
|
| 147 |
"learning_rate": 8.18911181775353e-06,
|
| 148 |
+
"loss": 0.0443,
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.3694581280788177,
|
| 153 |
+
"grad_norm": 0.046961464584805046,
|
| 154 |
"learning_rate": 7.93369417339209e-06,
|
| 155 |
+
"loss": 0.043,
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 0.3879310344827586,
|
| 160 |
+
"grad_norm": 0.03774079876533218,
|
| 161 |
"learning_rate": 7.666022164008458e-06,
|
| 162 |
+
"loss": 0.039,
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 0.4064039408866995,
|
| 167 |
+
"grad_norm": 0.039388091387549375,
|
| 168 |
"learning_rate": 7.387213885189746e-06,
|
| 169 |
+
"loss": 0.043,
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.4248768472906404,
|
| 174 |
+
"grad_norm": 0.04010524552891231,
|
| 175 |
"learning_rate": 7.098433949952146e-06,
|
| 176 |
+
"loss": 0.0418,
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.4433497536945813,
|
| 181 |
+
"grad_norm": 0.03364245597783716,
|
| 182 |
"learning_rate": 6.800888624023552e-06,
|
| 183 |
+
"loss": 0.0396,
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 0.4618226600985222,
|
| 188 |
+
"grad_norm": 0.03214650791918716,
|
| 189 |
"learning_rate": 6.495820787138209e-06,
|
| 190 |
+
"loss": 0.0343,
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.4802955665024631,
|
| 195 |
+
"grad_norm": 0.04248682387196562,
|
| 196 |
"learning_rate": 6.184504741390596e-06,
|
| 197 |
+
"loss": 0.0385,
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 0.4987684729064039,
|
| 202 |
+
"grad_norm": 0.041745690695414894,
|
| 203 |
"learning_rate": 5.8682408883346535e-06,
|
| 204 |
+
"loss": 0.0354,
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.5172413793103449,
|
| 209 |
+
"grad_norm": 0.043998119847084,
|
| 210 |
"learning_rate": 5.548350297062659e-06,
|
| 211 |
+
"loss": 0.0363,
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.5357142857142857,
|
| 216 |
+
"grad_norm": 0.03523911995673237,
|
| 217 |
"learning_rate": 5.2261691859535325e-06,
|
| 218 |
+
"loss": 0.0311,
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.5541871921182266,
|
| 223 |
+
"grad_norm": 0.051683301339415226,
|
| 224 |
"learning_rate": 4.903043341140879e-06,
|
| 225 |
+
"loss": 0.0322,
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 0.5726600985221675,
|
| 230 |
+
"grad_norm": 0.029645536386539162,
|
| 231 |
"learning_rate": 4.580322495015466e-06,
|
| 232 |
+
"loss": 0.0297,
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.5911330049261084,
|
| 237 |
+
"grad_norm": 0.038478089898929216,
|
| 238 |
"learning_rate": 4.259354688243758e-06,
|
| 239 |
+
"loss": 0.0373,
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 0.6096059113300493,
|
| 244 |
+
"grad_norm": 0.04016397060959619,
|
| 245 |
"learning_rate": 3.941480638852948e-06,
|
| 246 |
+
"loss": 0.0293,
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"epoch": 0.6280788177339901,
|
| 251 |
+
"grad_norm": 0.03295300026030036,
|
| 252 |
"learning_rate": 3.6280281419034934e-06,
|
| 253 |
+
"loss": 0.0306,
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 0.646551724137931,
|
| 258 |
+
"grad_norm": 0.04371251122688708,
|
| 259 |
"learning_rate": 3.3203065231422904e-06,
|
| 260 |
+
"loss": 0.0292,
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 0.6650246305418719,
|
| 265 |
+
"grad_norm": 0.02878432244430226,
|
| 266 |
"learning_rate": 3.019601169804216e-06,
|
| 267 |
+
"loss": 0.0348,
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
"epoch": 0.6834975369458128,
|
| 272 |
+
"grad_norm": 0.03582599925982462,
|
| 273 |
"learning_rate": 2.7271681614074973e-06,
|
| 274 |
+
"loss": 0.0292,
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 0.7019704433497537,
|
| 279 |
+
"grad_norm": 0.04498392884678493,
|
| 280 |
"learning_rate": 2.4442290229706344e-06,
|
| 281 |
+
"loss": 0.0355,
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 0.7204433497536946,
|
| 286 |
+
"grad_norm": 0.03888587884987569,
|
| 287 |
"learning_rate": 2.171965622567308e-06,
|
| 288 |
+
"loss": 0.0311,
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.7389162561576355,
|
| 293 |
+
"grad_norm": 0.04684421314626146,
|
| 294 |
"learning_rate": 1.9115152345327154e-06,
|
| 295 |
+
"loss": 0.0391,
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
"epoch": 0.7573891625615764,
|
| 300 |
+
"grad_norm": 0.031274018167510506,
|
| 301 |
"learning_rate": 1.6639657889429017e-06,
|
| 302 |
+
"loss": 0.0275,
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 0.7758620689655172,
|
| 307 |
+
"grad_norm": 0.03793766745215515,
|
| 308 |
"learning_rate": 1.4303513272105057e-06,
|
| 309 |
+
"loss": 0.0312,
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 0.7943349753694581,
|
| 314 |
+
"grad_norm": 0.03084526133895099,
|
| 315 |
"learning_rate": 1.2116476827794104e-06,
|
| 316 |
+
"loss": 0.0334,
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 0.812807881773399,
|
| 321 |
+
"grad_norm": 0.035909978135080484,
|
| 322 |
"learning_rate": 1.008768404960535e-06,
|
| 323 |
+
"loss": 0.0329,
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 0.8312807881773399,
|
| 328 |
+
"grad_norm": 0.03337500963867465,
|
| 329 |
"learning_rate": 8.225609429353187e-07,
|
| 330 |
+
"loss": 0.0299,
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 0.8497536945812808,
|
| 335 |
+
"grad_norm": 0.04194808091582502,
|
| 336 |
"learning_rate": 6.53803105866761e-07,
|
| 337 |
+
"loss": 0.0312,
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 0.8682266009852216,
|
| 342 |
+
"grad_norm": 0.030185202275612954,
|
| 343 |
"learning_rate": 5.031998139045352e-07,
|
| 344 |
+
"loss": 0.0291,
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.8866995073891626,
|
| 349 |
+
"grad_norm": 0.03678202209697851,
|
| 350 |
"learning_rate": 3.7138015365554834e-07,
|
| 351 |
+
"loss": 0.0329,
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
"epoch": 0.9051724137931034,
|
| 356 |
+
"grad_norm": 0.026908956915127933,
|
| 357 |
"learning_rate": 2.5889475041961767e-07,
|
| 358 |
+
"loss": 0.0318,
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 0.9236453201970444,
|
| 363 |
+
"grad_norm": 0.04036220391547687,
|
| 364 |
"learning_rate": 1.6621346816668993e-07,
|
| 365 |
+
"loss": 0.0304,
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
"epoch": 0.9421182266009852,
|
| 370 |
+
"grad_norm": 0.037246305929994866,
|
| 371 |
"learning_rate": 9.372344686307655e-08,
|
| 372 |
+
"loss": 0.036,
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
"epoch": 0.9605911330049262,
|
| 377 |
+
"grad_norm": 0.031641745281581694,
|
| 378 |
"learning_rate": 4.172748534499449e-08,
|
| 379 |
+
"loss": 0.0296,
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
"epoch": 0.979064039408867,
|
| 384 |
+
"grad_norm": 0.03135655037281048,
|
| 385 |
"learning_rate": 1.044277649433989e-08,
|
| 386 |
+
"loss": 0.0294,
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
"epoch": 0.9975369458128078,
|
| 391 |
+
"grad_norm": 0.03097346411772941,
|
| 392 |
"learning_rate": 0.0,
|
| 393 |
"loss": 0.033,
|
| 394 |
"step": 270
|
|
|
|
| 396 |
{
|
| 397 |
"epoch": 0.9975369458128078,
|
| 398 |
"step": 270,
|
| 399 |
+
"total_flos": 8.643970128528015e+17,
|
| 400 |
+
"train_loss": 0.07066047384783074,
|
| 401 |
+
"train_runtime": 3995.5402,
|
| 402 |
+
"train_samples_per_second": 1.625,
|
| 403 |
+
"train_steps_per_second": 0.068
|
| 404 |
}
|
| 405 |
],
|
| 406 |
"logging_steps": 5,
|
|
|
|
| 420 |
"attributes": {}
|
| 421 |
}
|
| 422 |
},
|
| 423 |
+
"total_flos": 8.643970128528015e+17,
|
| 424 |
"train_batch_size": 1,
|
| 425 |
"trial_name": null,
|
| 426 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 7352
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68d8b75a1101f455c3bf040fa2f84956565ddbcbbaa9e7e94c618c261996c857
|
| 3 |
size 7352
|