codeart-26m-ti-O3 / trainer_state.json
sheepy928's picture
Upload 11 files
cfad6f6 verified
{
"best_metric": 0.9297905882112695,
"best_model_checkpoint": "../save/marinda-default-ep0.4-26m-ti-di-only-O3-shuffle/checkpoint-600",
"epoch": 4.316546762589928,
"global_step": 600,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 9.971223021582734e-05,
"loss": 3.7929,
"step": 2
},
{
"epoch": 0.03,
"learning_rate": 9.942446043165468e-05,
"loss": 3.3681,
"step": 4
},
{
"epoch": 0.04,
"learning_rate": 9.913669064748203e-05,
"loss": 2.5519,
"step": 6
},
{
"epoch": 0.06,
"learning_rate": 9.884892086330935e-05,
"loss": 1.3021,
"step": 8
},
{
"epoch": 0.07,
"learning_rate": 9.856115107913669e-05,
"loss": 0.5662,
"step": 10
},
{
"epoch": 0.09,
"learning_rate": 9.827338129496403e-05,
"loss": 0.4006,
"step": 12
},
{
"epoch": 0.1,
"learning_rate": 9.798561151079138e-05,
"loss": 0.3735,
"step": 14
},
{
"epoch": 0.12,
"learning_rate": 9.769784172661871e-05,
"loss": 0.3599,
"step": 16
},
{
"epoch": 0.13,
"learning_rate": 9.741007194244605e-05,
"loss": 0.3238,
"step": 18
},
{
"epoch": 0.14,
"learning_rate": 9.712230215827339e-05,
"loss": 0.2998,
"step": 20
},
{
"epoch": 0.16,
"learning_rate": 9.683453237410072e-05,
"loss": 0.2738,
"step": 22
},
{
"epoch": 0.17,
"learning_rate": 9.654676258992806e-05,
"loss": 0.2698,
"step": 24
},
{
"epoch": 0.19,
"learning_rate": 9.62589928057554e-05,
"loss": 0.2616,
"step": 26
},
{
"epoch": 0.2,
"learning_rate": 9.597122302158273e-05,
"loss": 0.2386,
"step": 28
},
{
"epoch": 0.22,
"learning_rate": 9.568345323741009e-05,
"loss": 0.2266,
"step": 30
},
{
"epoch": 0.23,
"learning_rate": 9.539568345323742e-05,
"loss": 0.2159,
"step": 32
},
{
"epoch": 0.24,
"learning_rate": 9.510791366906475e-05,
"loss": 0.2166,
"step": 34
},
{
"epoch": 0.26,
"learning_rate": 9.482014388489208e-05,
"loss": 0.2089,
"step": 36
},
{
"epoch": 0.27,
"learning_rate": 9.453237410071943e-05,
"loss": 0.1988,
"step": 38
},
{
"epoch": 0.29,
"learning_rate": 9.424460431654677e-05,
"loss": 0.1958,
"step": 40
},
{
"epoch": 0.3,
"learning_rate": 9.39568345323741e-05,
"loss": 0.1898,
"step": 42
},
{
"epoch": 0.32,
"learning_rate": 9.366906474820144e-05,
"loss": 0.166,
"step": 44
},
{
"epoch": 0.33,
"learning_rate": 9.338129496402878e-05,
"loss": 0.1805,
"step": 46
},
{
"epoch": 0.35,
"learning_rate": 9.309352517985612e-05,
"loss": 0.1653,
"step": 48
},
{
"epoch": 0.36,
"learning_rate": 9.280575539568345e-05,
"loss": 0.155,
"step": 50
},
{
"epoch": 0.36,
"eval_f1": 0.3575783510975672,
"eval_loss": 0.13829465210437775,
"eval_precision": 0.42337897661762086,
"eval_recall": 0.30948700941378743,
"eval_runtime": 51.1038,
"eval_samples_per_second": 27.454,
"eval_steps_per_second": 0.117,
"step": 50
},
{
"epoch": 0.37,
"learning_rate": 9.251798561151079e-05,
"loss": 0.148,
"step": 52
},
{
"epoch": 0.39,
"learning_rate": 9.223021582733813e-05,
"loss": 0.1477,
"step": 54
},
{
"epoch": 0.4,
"learning_rate": 9.194244604316548e-05,
"loss": 0.1613,
"step": 56
},
{
"epoch": 0.42,
"learning_rate": 9.165467625899282e-05,
"loss": 0.1362,
"step": 58
},
{
"epoch": 0.43,
"learning_rate": 9.136690647482015e-05,
"loss": 0.1396,
"step": 60
},
{
"epoch": 0.45,
"learning_rate": 9.107913669064748e-05,
"loss": 0.136,
"step": 62
},
{
"epoch": 0.46,
"learning_rate": 9.079136690647483e-05,
"loss": 0.1397,
"step": 64
},
{
"epoch": 0.47,
"learning_rate": 9.050359712230216e-05,
"loss": 0.1337,
"step": 66
},
{
"epoch": 0.49,
"learning_rate": 9.02158273381295e-05,
"loss": 0.1232,
"step": 68
},
{
"epoch": 0.5,
"learning_rate": 8.992805755395684e-05,
"loss": 0.1229,
"step": 70
},
{
"epoch": 0.52,
"learning_rate": 8.964028776978419e-05,
"loss": 0.1284,
"step": 72
},
{
"epoch": 0.53,
"learning_rate": 8.935251798561151e-05,
"loss": 0.1198,
"step": 74
},
{
"epoch": 0.55,
"learning_rate": 8.906474820143885e-05,
"loss": 0.116,
"step": 76
},
{
"epoch": 0.56,
"learning_rate": 8.877697841726618e-05,
"loss": 0.1044,
"step": 78
},
{
"epoch": 0.58,
"learning_rate": 8.848920863309353e-05,
"loss": 0.0968,
"step": 80
},
{
"epoch": 0.59,
"learning_rate": 8.820143884892087e-05,
"loss": 0.1038,
"step": 82
},
{
"epoch": 0.6,
"learning_rate": 8.791366906474821e-05,
"loss": 0.0936,
"step": 84
},
{
"epoch": 0.62,
"learning_rate": 8.762589928057555e-05,
"loss": 0.117,
"step": 86
},
{
"epoch": 0.63,
"learning_rate": 8.733812949640288e-05,
"loss": 0.0876,
"step": 88
},
{
"epoch": 0.65,
"learning_rate": 8.705035971223022e-05,
"loss": 0.1114,
"step": 90
},
{
"epoch": 0.66,
"learning_rate": 8.676258992805756e-05,
"loss": 0.0882,
"step": 92
},
{
"epoch": 0.68,
"learning_rate": 8.647482014388489e-05,
"loss": 0.1021,
"step": 94
},
{
"epoch": 0.69,
"learning_rate": 8.618705035971223e-05,
"loss": 0.0864,
"step": 96
},
{
"epoch": 0.71,
"learning_rate": 8.589928057553958e-05,
"loss": 0.0885,
"step": 98
},
{
"epoch": 0.72,
"learning_rate": 8.561151079136692e-05,
"loss": 0.084,
"step": 100
},
{
"epoch": 0.72,
"eval_f1": 0.7231108774684605,
"eval_loss": 0.07253779470920563,
"eval_precision": 0.7317808420138814,
"eval_recall": 0.7146537125821346,
"eval_runtime": 51.1794,
"eval_samples_per_second": 27.413,
"eval_steps_per_second": 0.117,
"step": 100
},
{
"epoch": 0.73,
"learning_rate": 8.532374100719424e-05,
"loss": 0.0875,
"step": 102
},
{
"epoch": 0.75,
"learning_rate": 8.503597122302158e-05,
"loss": 0.0845,
"step": 104
},
{
"epoch": 0.76,
"learning_rate": 8.474820143884893e-05,
"loss": 0.0797,
"step": 106
},
{
"epoch": 0.78,
"learning_rate": 8.446043165467626e-05,
"loss": 0.082,
"step": 108
},
{
"epoch": 0.79,
"learning_rate": 8.41726618705036e-05,
"loss": 0.0933,
"step": 110
},
{
"epoch": 0.81,
"learning_rate": 8.388489208633094e-05,
"loss": 0.0815,
"step": 112
},
{
"epoch": 0.82,
"learning_rate": 8.359712230215828e-05,
"loss": 0.0814,
"step": 114
},
{
"epoch": 0.83,
"learning_rate": 8.330935251798561e-05,
"loss": 0.0715,
"step": 116
},
{
"epoch": 0.85,
"learning_rate": 8.302158273381295e-05,
"loss": 0.0728,
"step": 118
},
{
"epoch": 0.86,
"learning_rate": 8.273381294964029e-05,
"loss": 0.0788,
"step": 120
},
{
"epoch": 0.88,
"learning_rate": 8.244604316546764e-05,
"loss": 0.0718,
"step": 122
},
{
"epoch": 0.89,
"learning_rate": 8.215827338129497e-05,
"loss": 0.0717,
"step": 124
},
{
"epoch": 0.91,
"learning_rate": 8.187050359712231e-05,
"loss": 0.0718,
"step": 126
},
{
"epoch": 0.92,
"learning_rate": 8.158273381294965e-05,
"loss": 0.0767,
"step": 128
},
{
"epoch": 0.94,
"learning_rate": 8.129496402877698e-05,
"loss": 0.067,
"step": 130
},
{
"epoch": 0.95,
"learning_rate": 8.100719424460432e-05,
"loss": 0.0642,
"step": 132
},
{
"epoch": 0.96,
"learning_rate": 8.071942446043166e-05,
"loss": 0.0642,
"step": 134
},
{
"epoch": 0.98,
"learning_rate": 8.0431654676259e-05,
"loss": 0.072,
"step": 136
},
{
"epoch": 0.99,
"learning_rate": 8.014388489208633e-05,
"loss": 0.0625,
"step": 138
},
{
"epoch": 1.01,
"learning_rate": 7.985611510791368e-05,
"loss": 0.0681,
"step": 140
},
{
"epoch": 1.02,
"learning_rate": 7.9568345323741e-05,
"loss": 0.0558,
"step": 142
},
{
"epoch": 1.04,
"learning_rate": 7.928057553956834e-05,
"loss": 0.0708,
"step": 144
},
{
"epoch": 1.05,
"learning_rate": 7.899280575539568e-05,
"loss": 0.0664,
"step": 146
},
{
"epoch": 1.06,
"learning_rate": 7.870503597122303e-05,
"loss": 0.0657,
"step": 148
},
{
"epoch": 1.08,
"learning_rate": 7.841726618705037e-05,
"loss": 0.0602,
"step": 150
},
{
"epoch": 1.08,
"eval_f1": 0.8063984928986366,
"eval_loss": 0.053158584982156754,
"eval_precision": 0.7973809175439769,
"eval_recall": 0.8156325893754084,
"eval_runtime": 51.4131,
"eval_samples_per_second": 27.289,
"eval_steps_per_second": 0.117,
"step": 150
},
{
"epoch": 1.09,
"learning_rate": 7.81294964028777e-05,
"loss": 0.0672,
"step": 152
},
{
"epoch": 1.11,
"learning_rate": 7.784172661870504e-05,
"loss": 0.0622,
"step": 154
},
{
"epoch": 1.12,
"learning_rate": 7.755395683453238e-05,
"loss": 0.0525,
"step": 156
},
{
"epoch": 1.14,
"learning_rate": 7.726618705035971e-05,
"loss": 0.0646,
"step": 158
},
{
"epoch": 1.15,
"learning_rate": 7.697841726618705e-05,
"loss": 0.0619,
"step": 160
},
{
"epoch": 1.17,
"learning_rate": 7.669064748201439e-05,
"loss": 0.0518,
"step": 162
},
{
"epoch": 1.18,
"learning_rate": 7.640287769784174e-05,
"loss": 0.0538,
"step": 164
},
{
"epoch": 1.19,
"learning_rate": 7.611510791366907e-05,
"loss": 0.0604,
"step": 166
},
{
"epoch": 1.21,
"learning_rate": 7.582733812949641e-05,
"loss": 0.0563,
"step": 168
},
{
"epoch": 1.22,
"learning_rate": 7.553956834532374e-05,
"loss": 0.0503,
"step": 170
},
{
"epoch": 1.24,
"learning_rate": 7.525179856115109e-05,
"loss": 0.0524,
"step": 172
},
{
"epoch": 1.25,
"learning_rate": 7.496402877697842e-05,
"loss": 0.0618,
"step": 174
},
{
"epoch": 1.27,
"learning_rate": 7.467625899280576e-05,
"loss": 0.0516,
"step": 176
},
{
"epoch": 1.28,
"learning_rate": 7.43884892086331e-05,
"loss": 0.0516,
"step": 178
},
{
"epoch": 1.29,
"learning_rate": 7.410071942446043e-05,
"loss": 0.0488,
"step": 180
},
{
"epoch": 1.31,
"learning_rate": 7.381294964028777e-05,
"loss": 0.0686,
"step": 182
},
{
"epoch": 1.32,
"learning_rate": 7.352517985611511e-05,
"loss": 0.0534,
"step": 184
},
{
"epoch": 1.34,
"learning_rate": 7.323741007194244e-05,
"loss": 0.0548,
"step": 186
},
{
"epoch": 1.35,
"learning_rate": 7.294964028776978e-05,
"loss": 0.0554,
"step": 188
},
{
"epoch": 1.37,
"learning_rate": 7.266187050359713e-05,
"loss": 0.0547,
"step": 190
},
{
"epoch": 1.38,
"learning_rate": 7.237410071942447e-05,
"loss": 0.0516,
"step": 192
},
{
"epoch": 1.4,
"learning_rate": 7.20863309352518e-05,
"loss": 0.051,
"step": 194
},
{
"epoch": 1.41,
"learning_rate": 7.179856115107914e-05,
"loss": 0.0484,
"step": 196
},
{
"epoch": 1.42,
"learning_rate": 7.151079136690648e-05,
"loss": 0.053,
"step": 198
},
{
"epoch": 1.44,
"learning_rate": 7.122302158273382e-05,
"loss": 0.0469,
"step": 200
},
{
"epoch": 1.44,
"eval_f1": 0.8520792676649102,
"eval_loss": 0.03934021294116974,
"eval_precision": 0.8650739470116997,
"eval_recall": 0.8394789130496217,
"eval_runtime": 49.6143,
"eval_samples_per_second": 28.278,
"eval_steps_per_second": 0.121,
"step": 200
},
{
"epoch": 1.45,
"learning_rate": 7.093525179856115e-05,
"loss": 0.0585,
"step": 202
},
{
"epoch": 1.47,
"learning_rate": 7.064748201438849e-05,
"loss": 0.0441,
"step": 204
},
{
"epoch": 1.48,
"learning_rate": 7.035971223021584e-05,
"loss": 0.0511,
"step": 206
},
{
"epoch": 1.5,
"learning_rate": 7.007194244604318e-05,
"loss": 0.049,
"step": 208
},
{
"epoch": 1.51,
"learning_rate": 6.97841726618705e-05,
"loss": 0.049,
"step": 210
},
{
"epoch": 1.53,
"learning_rate": 6.949640287769784e-05,
"loss": 0.0469,
"step": 212
},
{
"epoch": 1.54,
"learning_rate": 6.920863309352519e-05,
"loss": 0.0473,
"step": 214
},
{
"epoch": 1.55,
"learning_rate": 6.892086330935252e-05,
"loss": 0.0418,
"step": 216
},
{
"epoch": 1.57,
"learning_rate": 6.863309352517986e-05,
"loss": 0.0491,
"step": 218
},
{
"epoch": 1.58,
"learning_rate": 6.83453237410072e-05,
"loss": 0.0519,
"step": 220
},
{
"epoch": 1.6,
"learning_rate": 6.805755395683453e-05,
"loss": 0.0465,
"step": 222
},
{
"epoch": 1.61,
"learning_rate": 6.776978417266187e-05,
"loss": 0.044,
"step": 224
},
{
"epoch": 1.63,
"learning_rate": 6.748201438848921e-05,
"loss": 0.0499,
"step": 226
},
{
"epoch": 1.64,
"learning_rate": 6.719424460431655e-05,
"loss": 0.0384,
"step": 228
},
{
"epoch": 1.65,
"learning_rate": 6.690647482014388e-05,
"loss": 0.048,
"step": 230
},
{
"epoch": 1.67,
"learning_rate": 6.661870503597123e-05,
"loss": 0.041,
"step": 232
},
{
"epoch": 1.68,
"learning_rate": 6.633093525179857e-05,
"loss": 0.0468,
"step": 234
},
{
"epoch": 1.7,
"learning_rate": 6.60431654676259e-05,
"loss": 0.0433,
"step": 236
},
{
"epoch": 1.71,
"learning_rate": 6.575539568345323e-05,
"loss": 0.041,
"step": 238
},
{
"epoch": 1.73,
"learning_rate": 6.546762589928058e-05,
"loss": 0.0455,
"step": 240
},
{
"epoch": 1.74,
"learning_rate": 6.517985611510792e-05,
"loss": 0.0462,
"step": 242
},
{
"epoch": 1.76,
"learning_rate": 6.489208633093525e-05,
"loss": 0.0438,
"step": 244
},
{
"epoch": 1.77,
"learning_rate": 6.460431654676259e-05,
"loss": 0.0459,
"step": 246
},
{
"epoch": 1.78,
"learning_rate": 6.431654676258994e-05,
"loss": 0.0428,
"step": 248
},
{
"epoch": 1.8,
"learning_rate": 6.402877697841726e-05,
"loss": 0.0457,
"step": 250
},
{
"epoch": 1.8,
"eval_f1": 0.8805310226478207,
"eval_loss": 0.03352304548025131,
"eval_precision": 0.8808927512662841,
"eval_recall": 0.880179582777523,
"eval_runtime": 52.5354,
"eval_samples_per_second": 26.706,
"eval_steps_per_second": 0.114,
"step": 250
},
{
"epoch": 1.81,
"learning_rate": 6.37410071942446e-05,
"loss": 0.044,
"step": 252
},
{
"epoch": 1.83,
"learning_rate": 6.345323741007194e-05,
"loss": 0.043,
"step": 254
},
{
"epoch": 1.84,
"learning_rate": 6.316546762589929e-05,
"loss": 0.0359,
"step": 256
},
{
"epoch": 1.86,
"learning_rate": 6.287769784172663e-05,
"loss": 0.0415,
"step": 258
},
{
"epoch": 1.87,
"learning_rate": 6.258992805755396e-05,
"loss": 0.0375,
"step": 260
},
{
"epoch": 1.88,
"learning_rate": 6.23021582733813e-05,
"loss": 0.0406,
"step": 262
},
{
"epoch": 1.9,
"learning_rate": 6.201438848920864e-05,
"loss": 0.033,
"step": 264
},
{
"epoch": 1.91,
"learning_rate": 6.172661870503597e-05,
"loss": 0.0444,
"step": 266
},
{
"epoch": 1.93,
"learning_rate": 6.143884892086331e-05,
"loss": 0.0448,
"step": 268
},
{
"epoch": 1.94,
"learning_rate": 6.115107913669065e-05,
"loss": 0.0381,
"step": 270
},
{
"epoch": 1.96,
"learning_rate": 6.0863309352517984e-05,
"loss": 0.0393,
"step": 272
},
{
"epoch": 1.97,
"learning_rate": 6.057553956834533e-05,
"loss": 0.0381,
"step": 274
},
{
"epoch": 1.99,
"learning_rate": 6.0287769784172665e-05,
"loss": 0.0359,
"step": 276
},
{
"epoch": 2.0,
"learning_rate": 6e-05,
"loss": 0.0407,
"step": 278
},
{
"epoch": 2.01,
"learning_rate": 5.971223021582734e-05,
"loss": 0.0337,
"step": 280
},
{
"epoch": 2.03,
"learning_rate": 5.942446043165468e-05,
"loss": 0.0413,
"step": 282
},
{
"epoch": 2.04,
"learning_rate": 5.913669064748202e-05,
"loss": 0.0394,
"step": 284
},
{
"epoch": 2.06,
"learning_rate": 5.8848920863309356e-05,
"loss": 0.0403,
"step": 286
},
{
"epoch": 2.07,
"learning_rate": 5.8561151079136686e-05,
"loss": 0.0436,
"step": 288
},
{
"epoch": 2.09,
"learning_rate": 5.8273381294964036e-05,
"loss": 0.04,
"step": 290
},
{
"epoch": 2.1,
"learning_rate": 5.798561151079137e-05,
"loss": 0.0374,
"step": 292
},
{
"epoch": 2.12,
"learning_rate": 5.7697841726618703e-05,
"loss": 0.0342,
"step": 294
},
{
"epoch": 2.13,
"learning_rate": 5.741007194244604e-05,
"loss": 0.03,
"step": 296
},
{
"epoch": 2.14,
"learning_rate": 5.712230215827339e-05,
"loss": 0.0365,
"step": 298
},
{
"epoch": 2.16,
"learning_rate": 5.683453237410072e-05,
"loss": 0.038,
"step": 300
},
{
"epoch": 2.16,
"eval_f1": 0.9002805338260745,
"eval_loss": 0.028384625911712646,
"eval_precision": 0.9072496256298583,
"eval_recall": 0.8934275403743082,
"eval_runtime": 50.0846,
"eval_samples_per_second": 28.013,
"eval_steps_per_second": 0.12,
"step": 300
},
{
"epoch": 2.17,
"learning_rate": 5.654676258992806e-05,
"loss": 0.0371,
"step": 302
},
{
"epoch": 2.19,
"learning_rate": 5.6258992805755395e-05,
"loss": 0.0364,
"step": 304
},
{
"epoch": 2.2,
"learning_rate": 5.597122302158274e-05,
"loss": 0.0375,
"step": 306
},
{
"epoch": 2.22,
"learning_rate": 5.5683453237410075e-05,
"loss": 0.0375,
"step": 308
},
{
"epoch": 2.23,
"learning_rate": 5.539568345323741e-05,
"loss": 0.0336,
"step": 310
},
{
"epoch": 2.24,
"learning_rate": 5.510791366906475e-05,
"loss": 0.0362,
"step": 312
},
{
"epoch": 2.26,
"learning_rate": 5.4820143884892086e-05,
"loss": 0.0307,
"step": 314
},
{
"epoch": 2.27,
"learning_rate": 5.453237410071943e-05,
"loss": 0.0359,
"step": 316
},
{
"epoch": 2.29,
"learning_rate": 5.4244604316546766e-05,
"loss": 0.037,
"step": 318
},
{
"epoch": 2.3,
"learning_rate": 5.39568345323741e-05,
"loss": 0.0311,
"step": 320
},
{
"epoch": 2.32,
"learning_rate": 5.366906474820143e-05,
"loss": 0.0345,
"step": 322
},
{
"epoch": 2.33,
"learning_rate": 5.3381294964028784e-05,
"loss": 0.026,
"step": 324
},
{
"epoch": 2.35,
"learning_rate": 5.309352517985612e-05,
"loss": 0.0323,
"step": 326
},
{
"epoch": 2.36,
"learning_rate": 5.280575539568345e-05,
"loss": 0.0329,
"step": 328
},
{
"epoch": 2.37,
"learning_rate": 5.251798561151079e-05,
"loss": 0.0368,
"step": 330
},
{
"epoch": 2.39,
"learning_rate": 5.223021582733814e-05,
"loss": 0.0381,
"step": 332
},
{
"epoch": 2.4,
"learning_rate": 5.194244604316547e-05,
"loss": 0.0351,
"step": 334
},
{
"epoch": 2.42,
"learning_rate": 5.1654676258992805e-05,
"loss": 0.0375,
"step": 336
},
{
"epoch": 2.43,
"learning_rate": 5.136690647482014e-05,
"loss": 0.0306,
"step": 338
},
{
"epoch": 2.45,
"learning_rate": 5.1079136690647486e-05,
"loss": 0.036,
"step": 340
},
{
"epoch": 2.46,
"learning_rate": 5.079136690647482e-05,
"loss": 0.0358,
"step": 342
},
{
"epoch": 2.47,
"learning_rate": 5.050359712230216e-05,
"loss": 0.0332,
"step": 344
},
{
"epoch": 2.49,
"learning_rate": 5.0215827338129496e-05,
"loss": 0.0353,
"step": 346
},
{
"epoch": 2.5,
"learning_rate": 4.992805755395683e-05,
"loss": 0.0349,
"step": 348
},
{
"epoch": 2.52,
"learning_rate": 4.964028776978418e-05,
"loss": 0.0374,
"step": 350
},
{
"epoch": 2.52,
"eval_f1": 0.9069143048629241,
"eval_loss": 0.02605625055730343,
"eval_precision": 0.908493352357095,
"eval_recall": 0.9053507022114148,
"eval_runtime": 50.6976,
"eval_samples_per_second": 27.674,
"eval_steps_per_second": 0.118,
"step": 350
},
{
"epoch": 2.53,
"learning_rate": 4.9352517985611514e-05,
"loss": 0.0271,
"step": 352
},
{
"epoch": 2.55,
"learning_rate": 4.906474820143885e-05,
"loss": 0.0305,
"step": 354
},
{
"epoch": 2.56,
"learning_rate": 4.877697841726619e-05,
"loss": 0.0337,
"step": 356
},
{
"epoch": 2.58,
"learning_rate": 4.848920863309353e-05,
"loss": 0.03,
"step": 358
},
{
"epoch": 2.59,
"learning_rate": 4.820143884892087e-05,
"loss": 0.0291,
"step": 360
},
{
"epoch": 2.6,
"learning_rate": 4.79136690647482e-05,
"loss": 0.0255,
"step": 362
},
{
"epoch": 2.62,
"learning_rate": 4.762589928057554e-05,
"loss": 0.0273,
"step": 364
},
{
"epoch": 2.63,
"learning_rate": 4.733812949640288e-05,
"loss": 0.0354,
"step": 366
},
{
"epoch": 2.65,
"learning_rate": 4.7050359712230216e-05,
"loss": 0.0357,
"step": 368
},
{
"epoch": 2.66,
"learning_rate": 4.676258992805755e-05,
"loss": 0.0326,
"step": 370
},
{
"epoch": 2.68,
"learning_rate": 4.6474820143884896e-05,
"loss": 0.0336,
"step": 372
},
{
"epoch": 2.69,
"learning_rate": 4.618705035971223e-05,
"loss": 0.0327,
"step": 374
},
{
"epoch": 2.71,
"learning_rate": 4.589928057553957e-05,
"loss": 0.0305,
"step": 376
},
{
"epoch": 2.72,
"learning_rate": 4.561151079136691e-05,
"loss": 0.031,
"step": 378
},
{
"epoch": 2.73,
"learning_rate": 4.532374100719425e-05,
"loss": 0.0328,
"step": 380
},
{
"epoch": 2.75,
"learning_rate": 4.503597122302158e-05,
"loss": 0.0316,
"step": 382
},
{
"epoch": 2.76,
"learning_rate": 4.4748201438848924e-05,
"loss": 0.0328,
"step": 384
},
{
"epoch": 2.78,
"learning_rate": 4.446043165467626e-05,
"loss": 0.0281,
"step": 386
},
{
"epoch": 2.79,
"learning_rate": 4.41726618705036e-05,
"loss": 0.0317,
"step": 388
},
{
"epoch": 2.81,
"learning_rate": 4.3884892086330935e-05,
"loss": 0.0312,
"step": 390
},
{
"epoch": 2.82,
"learning_rate": 4.359712230215828e-05,
"loss": 0.0285,
"step": 392
},
{
"epoch": 2.83,
"learning_rate": 4.3309352517985616e-05,
"loss": 0.0305,
"step": 394
},
{
"epoch": 2.85,
"learning_rate": 4.302158273381295e-05,
"loss": 0.0282,
"step": 396
},
{
"epoch": 2.86,
"learning_rate": 4.273381294964029e-05,
"loss": 0.0242,
"step": 398
},
{
"epoch": 2.88,
"learning_rate": 4.244604316546763e-05,
"loss": 0.0311,
"step": 400
},
{
"epoch": 2.88,
"eval_f1": 0.9081299492825968,
"eval_loss": 0.025341803207993507,
"eval_precision": 0.9010941229613114,
"eval_recall": 0.9152866704090037,
"eval_runtime": 50.33,
"eval_samples_per_second": 27.876,
"eval_steps_per_second": 0.119,
"step": 400
},
{
"epoch": 2.89,
"learning_rate": 4.215827338129496e-05,
"loss": 0.0296,
"step": 402
},
{
"epoch": 2.91,
"learning_rate": 4.18705035971223e-05,
"loss": 0.0324,
"step": 404
},
{
"epoch": 2.92,
"learning_rate": 4.1582733812949644e-05,
"loss": 0.033,
"step": 406
},
{
"epoch": 2.94,
"learning_rate": 4.129496402877698e-05,
"loss": 0.0313,
"step": 408
},
{
"epoch": 2.95,
"learning_rate": 4.100719424460432e-05,
"loss": 0.0244,
"step": 410
},
{
"epoch": 2.96,
"learning_rate": 4.0719424460431654e-05,
"loss": 0.0285,
"step": 412
},
{
"epoch": 2.98,
"learning_rate": 4.0431654676259e-05,
"loss": 0.0363,
"step": 414
},
{
"epoch": 2.99,
"learning_rate": 4.014388489208633e-05,
"loss": 0.0281,
"step": 416
},
{
"epoch": 3.01,
"learning_rate": 3.985611510791367e-05,
"loss": 0.0326,
"step": 418
},
{
"epoch": 3.02,
"learning_rate": 3.956834532374101e-05,
"loss": 0.0253,
"step": 420
},
{
"epoch": 3.04,
"learning_rate": 3.9280575539568345e-05,
"loss": 0.0265,
"step": 422
},
{
"epoch": 3.05,
"learning_rate": 3.899280575539568e-05,
"loss": 0.0346,
"step": 424
},
{
"epoch": 3.06,
"learning_rate": 3.8705035971223026e-05,
"loss": 0.0286,
"step": 426
},
{
"epoch": 3.08,
"learning_rate": 3.841726618705036e-05,
"loss": 0.0251,
"step": 428
},
{
"epoch": 3.09,
"learning_rate": 3.81294964028777e-05,
"loss": 0.0266,
"step": 430
},
{
"epoch": 3.11,
"learning_rate": 3.784172661870504e-05,
"loss": 0.0308,
"step": 432
},
{
"epoch": 3.12,
"learning_rate": 3.755395683453238e-05,
"loss": 0.0268,
"step": 434
},
{
"epoch": 3.14,
"learning_rate": 3.726618705035971e-05,
"loss": 0.0255,
"step": 436
},
{
"epoch": 3.15,
"learning_rate": 3.6978417266187054e-05,
"loss": 0.028,
"step": 438
},
{
"epoch": 3.17,
"learning_rate": 3.669064748201439e-05,
"loss": 0.0293,
"step": 440
},
{
"epoch": 3.18,
"learning_rate": 3.640287769784173e-05,
"loss": 0.0214,
"step": 442
},
{
"epoch": 3.19,
"learning_rate": 3.6115107913669065e-05,
"loss": 0.0248,
"step": 444
},
{
"epoch": 3.21,
"learning_rate": 3.582733812949641e-05,
"loss": 0.026,
"step": 446
},
{
"epoch": 3.22,
"learning_rate": 3.5539568345323745e-05,
"loss": 0.0287,
"step": 448
},
{
"epoch": 3.24,
"learning_rate": 3.5251798561151075e-05,
"loss": 0.025,
"step": 450
},
{
"epoch": 3.24,
"eval_f1": 0.9162968278365243,
"eval_loss": 0.022837216034531593,
"eval_precision": 0.9160659107576388,
"eval_recall": 0.9165378664042556,
"eval_runtime": 49.8961,
"eval_samples_per_second": 28.118,
"eval_steps_per_second": 0.12,
"step": 450
},
{
"epoch": 3.25,
"learning_rate": 3.496402877697842e-05,
"loss": 0.0217,
"step": 452
},
{
"epoch": 3.27,
"learning_rate": 3.4676258992805756e-05,
"loss": 0.0274,
"step": 454
},
{
"epoch": 3.28,
"learning_rate": 3.438848920863309e-05,
"loss": 0.0172,
"step": 456
},
{
"epoch": 3.29,
"learning_rate": 3.410071942446043e-05,
"loss": 0.032,
"step": 458
},
{
"epoch": 3.31,
"learning_rate": 3.3812949640287773e-05,
"loss": 0.0251,
"step": 460
},
{
"epoch": 3.32,
"learning_rate": 3.352517985611511e-05,
"loss": 0.0214,
"step": 462
},
{
"epoch": 3.34,
"learning_rate": 3.323741007194245e-05,
"loss": 0.0248,
"step": 464
},
{
"epoch": 3.35,
"learning_rate": 3.2949640287769784e-05,
"loss": 0.0291,
"step": 466
},
{
"epoch": 3.37,
"learning_rate": 3.266187050359713e-05,
"loss": 0.0259,
"step": 468
},
{
"epoch": 3.38,
"learning_rate": 3.237410071942446e-05,
"loss": 0.0247,
"step": 470
},
{
"epoch": 3.4,
"learning_rate": 3.20863309352518e-05,
"loss": 0.0293,
"step": 472
},
{
"epoch": 3.41,
"learning_rate": 3.179856115107914e-05,
"loss": 0.0331,
"step": 474
},
{
"epoch": 3.42,
"learning_rate": 3.1510791366906475e-05,
"loss": 0.026,
"step": 476
},
{
"epoch": 3.44,
"learning_rate": 3.122302158273381e-05,
"loss": 0.0279,
"step": 478
},
{
"epoch": 3.45,
"learning_rate": 3.0935251798561156e-05,
"loss": 0.0263,
"step": 480
},
{
"epoch": 3.47,
"learning_rate": 3.064748201438849e-05,
"loss": 0.0303,
"step": 482
},
{
"epoch": 3.48,
"learning_rate": 3.035971223021583e-05,
"loss": 0.0276,
"step": 484
},
{
"epoch": 3.5,
"learning_rate": 3.0071942446043167e-05,
"loss": 0.0284,
"step": 486
},
{
"epoch": 3.51,
"learning_rate": 2.9784172661870507e-05,
"loss": 0.0278,
"step": 488
},
{
"epoch": 3.53,
"learning_rate": 2.9496402877697844e-05,
"loss": 0.0248,
"step": 490
},
{
"epoch": 3.54,
"learning_rate": 2.9208633093525184e-05,
"loss": 0.0249,
"step": 492
},
{
"epoch": 3.55,
"learning_rate": 2.892086330935252e-05,
"loss": 0.0275,
"step": 494
},
{
"epoch": 3.57,
"learning_rate": 2.8633093525179854e-05,
"loss": 0.0267,
"step": 496
},
{
"epoch": 3.58,
"learning_rate": 2.8345323741007195e-05,
"loss": 0.0253,
"step": 498
},
{
"epoch": 3.6,
"learning_rate": 2.805755395683453e-05,
"loss": 0.0245,
"step": 500
},
{
"epoch": 3.6,
"eval_f1": 0.9240470400791703,
"eval_loss": 0.021140407770872116,
"eval_precision": 0.9307795691974465,
"eval_recall": 0.9174210635773746,
"eval_runtime": 50.9677,
"eval_samples_per_second": 27.527,
"eval_steps_per_second": 0.118,
"step": 500
},
{
"epoch": 3.61,
"learning_rate": 2.7769784172661872e-05,
"loss": 0.0234,
"step": 502
},
{
"epoch": 3.63,
"learning_rate": 2.748201438848921e-05,
"loss": 0.0304,
"step": 504
},
{
"epoch": 3.64,
"learning_rate": 2.719424460431655e-05,
"loss": 0.0254,
"step": 506
},
{
"epoch": 3.65,
"learning_rate": 2.6906474820143886e-05,
"loss": 0.0279,
"step": 508
},
{
"epoch": 3.67,
"learning_rate": 2.6618705035971226e-05,
"loss": 0.0273,
"step": 510
},
{
"epoch": 3.68,
"learning_rate": 2.633093525179856e-05,
"loss": 0.0263,
"step": 512
},
{
"epoch": 3.7,
"learning_rate": 2.6043165467625903e-05,
"loss": 0.0257,
"step": 514
},
{
"epoch": 3.71,
"learning_rate": 2.5755395683453237e-05,
"loss": 0.024,
"step": 516
},
{
"epoch": 3.73,
"learning_rate": 2.5467625899280577e-05,
"loss": 0.0215,
"step": 518
},
{
"epoch": 3.74,
"learning_rate": 2.5179856115107914e-05,
"loss": 0.0232,
"step": 520
},
{
"epoch": 3.76,
"learning_rate": 2.489208633093525e-05,
"loss": 0.0267,
"step": 522
},
{
"epoch": 3.77,
"learning_rate": 2.460431654676259e-05,
"loss": 0.0328,
"step": 524
},
{
"epoch": 3.78,
"learning_rate": 2.4316546762589928e-05,
"loss": 0.0236,
"step": 526
},
{
"epoch": 3.8,
"learning_rate": 2.4028776978417268e-05,
"loss": 0.0237,
"step": 528
},
{
"epoch": 3.81,
"learning_rate": 2.3741007194244605e-05,
"loss": 0.0254,
"step": 530
},
{
"epoch": 3.83,
"learning_rate": 2.3453237410071942e-05,
"loss": 0.0225,
"step": 532
},
{
"epoch": 3.84,
"learning_rate": 2.3165467625899282e-05,
"loss": 0.0285,
"step": 534
},
{
"epoch": 3.86,
"learning_rate": 2.287769784172662e-05,
"loss": 0.0283,
"step": 536
},
{
"epoch": 3.87,
"learning_rate": 2.258992805755396e-05,
"loss": 0.0233,
"step": 538
},
{
"epoch": 3.88,
"learning_rate": 2.2302158273381296e-05,
"loss": 0.0223,
"step": 540
},
{
"epoch": 3.9,
"learning_rate": 2.2014388489208633e-05,
"loss": 0.0257,
"step": 542
},
{
"epoch": 3.91,
"learning_rate": 2.1726618705035974e-05,
"loss": 0.0255,
"step": 544
},
{
"epoch": 3.93,
"learning_rate": 2.143884892086331e-05,
"loss": 0.0215,
"step": 546
},
{
"epoch": 3.94,
"learning_rate": 2.115107913669065e-05,
"loss": 0.021,
"step": 548
},
{
"epoch": 3.96,
"learning_rate": 2.0863309352517988e-05,
"loss": 0.0239,
"step": 550
},
{
"epoch": 3.96,
"eval_f1": 0.9255869648252579,
"eval_loss": 0.020801428705453873,
"eval_precision": 0.9219423140402028,
"eval_recall": 0.9292706256500548,
"eval_runtime": 48.2186,
"eval_samples_per_second": 29.097,
"eval_steps_per_second": 0.124,
"step": 550
},
{
"epoch": 3.97,
"learning_rate": 2.0575539568345324e-05,
"loss": 0.0226,
"step": 552
},
{
"epoch": 3.99,
"learning_rate": 2.0287769784172665e-05,
"loss": 0.0217,
"step": 554
},
{
"epoch": 4.0,
"learning_rate": 2e-05,
"loss": 0.0279,
"step": 556
},
{
"epoch": 4.01,
"learning_rate": 1.9712230215827342e-05,
"loss": 0.0271,
"step": 558
},
{
"epoch": 4.03,
"learning_rate": 1.942446043165468e-05,
"loss": 0.0235,
"step": 560
},
{
"epoch": 4.04,
"learning_rate": 1.9136690647482016e-05,
"loss": 0.0239,
"step": 562
},
{
"epoch": 4.06,
"learning_rate": 1.8848920863309353e-05,
"loss": 0.0235,
"step": 564
},
{
"epoch": 4.07,
"learning_rate": 1.856115107913669e-05,
"loss": 0.0231,
"step": 566
},
{
"epoch": 4.09,
"learning_rate": 1.827338129496403e-05,
"loss": 0.0286,
"step": 568
},
{
"epoch": 4.1,
"learning_rate": 1.7985611510791367e-05,
"loss": 0.0225,
"step": 570
},
{
"epoch": 4.12,
"learning_rate": 1.7697841726618707e-05,
"loss": 0.0272,
"step": 572
},
{
"epoch": 4.13,
"learning_rate": 1.7410071942446044e-05,
"loss": 0.0235,
"step": 574
},
{
"epoch": 4.14,
"learning_rate": 1.712230215827338e-05,
"loss": 0.0243,
"step": 576
},
{
"epoch": 4.16,
"learning_rate": 1.683453237410072e-05,
"loss": 0.0209,
"step": 578
},
{
"epoch": 4.17,
"learning_rate": 1.6546762589928058e-05,
"loss": 0.0253,
"step": 580
},
{
"epoch": 4.19,
"learning_rate": 1.6258992805755395e-05,
"loss": 0.0199,
"step": 582
},
{
"epoch": 4.2,
"learning_rate": 1.5971223021582735e-05,
"loss": 0.0223,
"step": 584
},
{
"epoch": 4.22,
"learning_rate": 1.5683453237410072e-05,
"loss": 0.0214,
"step": 586
},
{
"epoch": 4.23,
"learning_rate": 1.5395683453237412e-05,
"loss": 0.02,
"step": 588
},
{
"epoch": 4.24,
"learning_rate": 1.5107913669064749e-05,
"loss": 0.024,
"step": 590
},
{
"epoch": 4.26,
"learning_rate": 1.4820143884892088e-05,
"loss": 0.0267,
"step": 592
},
{
"epoch": 4.27,
"learning_rate": 1.4532374100719426e-05,
"loss": 0.0234,
"step": 594
},
{
"epoch": 4.29,
"learning_rate": 1.4244604316546765e-05,
"loss": 0.0263,
"step": 596
},
{
"epoch": 4.3,
"learning_rate": 1.3956834532374102e-05,
"loss": 0.0203,
"step": 598
},
{
"epoch": 4.32,
"learning_rate": 1.366906474820144e-05,
"loss": 0.0218,
"step": 600
},
{
"epoch": 4.32,
"eval_f1": 0.9297905882112695,
"eval_loss": 0.020102111622691154,
"eval_precision": 0.9272434483038768,
"eval_recall": 0.9323618157559713,
"eval_runtime": 47.4996,
"eval_samples_per_second": 29.537,
"eval_steps_per_second": 0.126,
"step": 600
}
],
"max_steps": 695,
"num_train_epochs": 5,
"total_flos": 3.00872626077696e+16,
"trial_name": null,
"trial_params": null
}