dimasik2987's picture
Training in progress, step 75, checkpoint
c83df69 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.08605851979345955,
"eval_steps": 25,
"global_step": 75,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0011474469305794606,
"grad_norm": 0.11429613083600998,
"learning_rate": 3.3333333333333335e-05,
"loss": 0.0272,
"step": 1
},
{
"epoch": 0.0011474469305794606,
"eval_loss": 0.32320111989974976,
"eval_runtime": 54.0894,
"eval_samples_per_second": 13.57,
"eval_steps_per_second": 6.785,
"step": 1
},
{
"epoch": 0.002294893861158921,
"grad_norm": 0.13278168439865112,
"learning_rate": 6.666666666666667e-05,
"loss": 0.0323,
"step": 2
},
{
"epoch": 0.0034423407917383822,
"grad_norm": 0.13376079499721527,
"learning_rate": 0.0001,
"loss": 0.0362,
"step": 3
},
{
"epoch": 0.004589787722317842,
"grad_norm": 0.12864063680171967,
"learning_rate": 9.99524110790929e-05,
"loss": 0.0436,
"step": 4
},
{
"epoch": 0.005737234652897304,
"grad_norm": 0.12688331305980682,
"learning_rate": 9.980973490458728e-05,
"loss": 0.0377,
"step": 5
},
{
"epoch": 0.0068846815834767644,
"grad_norm": 0.15226823091506958,
"learning_rate": 9.957224306869053e-05,
"loss": 0.038,
"step": 6
},
{
"epoch": 0.008032128514056224,
"grad_norm": 0.17574366927146912,
"learning_rate": 9.924038765061042e-05,
"loss": 0.0409,
"step": 7
},
{
"epoch": 0.009179575444635685,
"grad_norm": 0.1377338469028473,
"learning_rate": 9.881480035599667e-05,
"loss": 0.0419,
"step": 8
},
{
"epoch": 0.010327022375215147,
"grad_norm": 0.16289022564888,
"learning_rate": 9.829629131445342e-05,
"loss": 0.0468,
"step": 9
},
{
"epoch": 0.011474469305794608,
"grad_norm": 0.2999342083930969,
"learning_rate": 9.768584753741134e-05,
"loss": 0.0619,
"step": 10
},
{
"epoch": 0.012621916236374068,
"grad_norm": 0.2969723343849182,
"learning_rate": 9.698463103929542e-05,
"loss": 0.0559,
"step": 11
},
{
"epoch": 0.013769363166953529,
"grad_norm": 0.20262478291988373,
"learning_rate": 9.619397662556435e-05,
"loss": 0.0371,
"step": 12
},
{
"epoch": 0.01491681009753299,
"grad_norm": 0.22081464529037476,
"learning_rate": 9.53153893518325e-05,
"loss": 0.0369,
"step": 13
},
{
"epoch": 0.01606425702811245,
"grad_norm": 0.24801427125930786,
"learning_rate": 9.435054165891109e-05,
"loss": 0.0348,
"step": 14
},
{
"epoch": 0.01721170395869191,
"grad_norm": 0.2353007048368454,
"learning_rate": 9.330127018922194e-05,
"loss": 0.0364,
"step": 15
},
{
"epoch": 0.01835915088927137,
"grad_norm": 0.2555350661277771,
"learning_rate": 9.21695722906443e-05,
"loss": 0.032,
"step": 16
},
{
"epoch": 0.019506597819850834,
"grad_norm": 0.2861219644546509,
"learning_rate": 9.09576022144496e-05,
"loss": 0.0344,
"step": 17
},
{
"epoch": 0.020654044750430294,
"grad_norm": 0.3375321626663208,
"learning_rate": 8.966766701456177e-05,
"loss": 0.0356,
"step": 18
},
{
"epoch": 0.021801491681009755,
"grad_norm": 0.24729681015014648,
"learning_rate": 8.83022221559489e-05,
"loss": 0.0227,
"step": 19
},
{
"epoch": 0.022948938611589215,
"grad_norm": 0.2967877686023712,
"learning_rate": 8.68638668405062e-05,
"loss": 0.0254,
"step": 20
},
{
"epoch": 0.024096385542168676,
"grad_norm": 0.2003224939107895,
"learning_rate": 8.535533905932738e-05,
"loss": 0.0142,
"step": 21
},
{
"epoch": 0.025243832472748137,
"grad_norm": 0.5681616067886353,
"learning_rate": 8.377951038078302e-05,
"loss": 0.0188,
"step": 22
},
{
"epoch": 0.026391279403327597,
"grad_norm": 0.3567965030670166,
"learning_rate": 8.213938048432697e-05,
"loss": 0.028,
"step": 23
},
{
"epoch": 0.027538726333907058,
"grad_norm": 0.6463397145271301,
"learning_rate": 8.043807145043604e-05,
"loss": 0.0238,
"step": 24
},
{
"epoch": 0.02868617326448652,
"grad_norm": 0.24729986488819122,
"learning_rate": 7.86788218175523e-05,
"loss": 0.0193,
"step": 25
},
{
"epoch": 0.02868617326448652,
"eval_loss": 0.04240313917398453,
"eval_runtime": 54.5762,
"eval_samples_per_second": 13.449,
"eval_steps_per_second": 6.725,
"step": 25
},
{
"epoch": 0.02983362019506598,
"grad_norm": 0.43531009554862976,
"learning_rate": 7.68649804173412e-05,
"loss": 0.0243,
"step": 26
},
{
"epoch": 0.03098106712564544,
"grad_norm": 0.48705872893333435,
"learning_rate": 7.500000000000001e-05,
"loss": 0.0254,
"step": 27
},
{
"epoch": 0.0321285140562249,
"grad_norm": 0.42778128385543823,
"learning_rate": 7.308743066175172e-05,
"loss": 0.0228,
"step": 28
},
{
"epoch": 0.03327596098680436,
"grad_norm": 0.5449861884117126,
"learning_rate": 7.113091308703498e-05,
"loss": 0.0421,
"step": 29
},
{
"epoch": 0.03442340791738382,
"grad_norm": 0.3300516903400421,
"learning_rate": 6.91341716182545e-05,
"loss": 0.0217,
"step": 30
},
{
"epoch": 0.03557085484796328,
"grad_norm": 0.36909106373786926,
"learning_rate": 6.710100716628344e-05,
"loss": 0.0211,
"step": 31
},
{
"epoch": 0.03671830177854274,
"grad_norm": 0.32208243012428284,
"learning_rate": 6.503528997521366e-05,
"loss": 0.0221,
"step": 32
},
{
"epoch": 0.0378657487091222,
"grad_norm": 0.5139682292938232,
"learning_rate": 6.294095225512603e-05,
"loss": 0.0351,
"step": 33
},
{
"epoch": 0.03901319563970167,
"grad_norm": 0.4424279034137726,
"learning_rate": 6.0821980696905146e-05,
"loss": 0.0402,
"step": 34
},
{
"epoch": 0.040160642570281124,
"grad_norm": 0.406888872385025,
"learning_rate": 5.868240888334653e-05,
"loss": 0.0256,
"step": 35
},
{
"epoch": 0.04130808950086059,
"grad_norm": 0.3193250298500061,
"learning_rate": 5.6526309611002594e-05,
"loss": 0.0306,
"step": 36
},
{
"epoch": 0.042455536431440045,
"grad_norm": 0.5056396126747131,
"learning_rate": 5.435778713738292e-05,
"loss": 0.0446,
"step": 37
},
{
"epoch": 0.04360298336201951,
"grad_norm": 0.4755103886127472,
"learning_rate": 5.218096936826681e-05,
"loss": 0.0345,
"step": 38
},
{
"epoch": 0.04475043029259897,
"grad_norm": 0.41109639406204224,
"learning_rate": 5e-05,
"loss": 0.0402,
"step": 39
},
{
"epoch": 0.04589787722317843,
"grad_norm": 0.48927897214889526,
"learning_rate": 4.781903063173321e-05,
"loss": 0.0431,
"step": 40
},
{
"epoch": 0.04704532415375789,
"grad_norm": 0.4970654249191284,
"learning_rate": 4.564221286261709e-05,
"loss": 0.0468,
"step": 41
},
{
"epoch": 0.04819277108433735,
"grad_norm": 0.3974848687648773,
"learning_rate": 4.347369038899744e-05,
"loss": 0.0347,
"step": 42
},
{
"epoch": 0.04934021801491681,
"grad_norm": 0.5659162402153015,
"learning_rate": 4.131759111665349e-05,
"loss": 0.0475,
"step": 43
},
{
"epoch": 0.05048766494549627,
"grad_norm": 0.4630831778049469,
"learning_rate": 3.917801930309486e-05,
"loss": 0.0555,
"step": 44
},
{
"epoch": 0.05163511187607573,
"grad_norm": 0.6687282919883728,
"learning_rate": 3.705904774487396e-05,
"loss": 0.0602,
"step": 45
},
{
"epoch": 0.052782558806655194,
"grad_norm": 0.5463370680809021,
"learning_rate": 3.4964710024786354e-05,
"loss": 0.0574,
"step": 46
},
{
"epoch": 0.05393000573723465,
"grad_norm": 0.6840320825576782,
"learning_rate": 3.289899283371657e-05,
"loss": 0.074,
"step": 47
},
{
"epoch": 0.055077452667814115,
"grad_norm": 0.7155039310455322,
"learning_rate": 3.086582838174551e-05,
"loss": 0.0904,
"step": 48
},
{
"epoch": 0.05622489959839357,
"grad_norm": 2.343247652053833,
"learning_rate": 2.886908691296504e-05,
"loss": 0.1657,
"step": 49
},
{
"epoch": 0.05737234652897304,
"grad_norm": 5.246421813964844,
"learning_rate": 2.6912569338248315e-05,
"loss": 0.3323,
"step": 50
},
{
"epoch": 0.05737234652897304,
"eval_loss": 0.03461047261953354,
"eval_runtime": 54.72,
"eval_samples_per_second": 13.414,
"eval_steps_per_second": 6.707,
"step": 50
},
{
"epoch": 0.058519793459552494,
"grad_norm": 0.1377524882555008,
"learning_rate": 2.500000000000001e-05,
"loss": 0.0194,
"step": 51
},
{
"epoch": 0.05966724039013196,
"grad_norm": 0.18886782228946686,
"learning_rate": 2.3135019582658802e-05,
"loss": 0.0154,
"step": 52
},
{
"epoch": 0.060814687320711415,
"grad_norm": 0.37080180644989014,
"learning_rate": 2.132117818244771e-05,
"loss": 0.017,
"step": 53
},
{
"epoch": 0.06196213425129088,
"grad_norm": 0.19527259469032288,
"learning_rate": 1.9561928549563968e-05,
"loss": 0.0126,
"step": 54
},
{
"epoch": 0.06310958118187034,
"grad_norm": 0.18304114043712616,
"learning_rate": 1.7860619515673033e-05,
"loss": 0.0188,
"step": 55
},
{
"epoch": 0.0642570281124498,
"grad_norm": 0.16556933522224426,
"learning_rate": 1.622048961921699e-05,
"loss": 0.0117,
"step": 56
},
{
"epoch": 0.06540447504302926,
"grad_norm": 0.14504587650299072,
"learning_rate": 1.4644660940672627e-05,
"loss": 0.016,
"step": 57
},
{
"epoch": 0.06655192197360872,
"grad_norm": 0.13935603201389313,
"learning_rate": 1.3136133159493802e-05,
"loss": 0.0116,
"step": 58
},
{
"epoch": 0.06769936890418818,
"grad_norm": 0.18318995833396912,
"learning_rate": 1.1697777844051105e-05,
"loss": 0.021,
"step": 59
},
{
"epoch": 0.06884681583476764,
"grad_norm": 0.17003515362739563,
"learning_rate": 1.0332332985438248e-05,
"loss": 0.018,
"step": 60
},
{
"epoch": 0.0699942627653471,
"grad_norm": 0.16662214696407318,
"learning_rate": 9.042397785550405e-06,
"loss": 0.0149,
"step": 61
},
{
"epoch": 0.07114170969592656,
"grad_norm": 0.2166244387626648,
"learning_rate": 7.830427709355725e-06,
"loss": 0.0115,
"step": 62
},
{
"epoch": 0.07228915662650602,
"grad_norm": 0.2247883826494217,
"learning_rate": 6.698729810778065e-06,
"loss": 0.0275,
"step": 63
},
{
"epoch": 0.07343660355708548,
"grad_norm": 0.19006967544555664,
"learning_rate": 5.649458341088915e-06,
"loss": 0.0186,
"step": 64
},
{
"epoch": 0.07458405048766495,
"grad_norm": 0.18058659136295319,
"learning_rate": 4.684610648167503e-06,
"loss": 0.0156,
"step": 65
},
{
"epoch": 0.0757314974182444,
"grad_norm": 0.24158023297786713,
"learning_rate": 3.8060233744356633e-06,
"loss": 0.02,
"step": 66
},
{
"epoch": 0.07687894434882386,
"grad_norm": 0.25969210267066956,
"learning_rate": 3.0153689607045845e-06,
"loss": 0.0192,
"step": 67
},
{
"epoch": 0.07802639127940333,
"grad_norm": 0.254467248916626,
"learning_rate": 2.314152462588659e-06,
"loss": 0.0215,
"step": 68
},
{
"epoch": 0.07917383820998279,
"grad_norm": 0.1745733618736267,
"learning_rate": 1.70370868554659e-06,
"loss": 0.0125,
"step": 69
},
{
"epoch": 0.08032128514056225,
"grad_norm": 0.2833689749240875,
"learning_rate": 1.1851996440033319e-06,
"loss": 0.0258,
"step": 70
},
{
"epoch": 0.0814687320711417,
"grad_norm": 0.17747850716114044,
"learning_rate": 7.596123493895991e-07,
"loss": 0.0168,
"step": 71
},
{
"epoch": 0.08261617900172118,
"grad_norm": 0.2795718312263489,
"learning_rate": 4.277569313094809e-07,
"loss": 0.0238,
"step": 72
},
{
"epoch": 0.08376362593230063,
"grad_norm": 0.3440586030483246,
"learning_rate": 1.9026509541272275e-07,
"loss": 0.0383,
"step": 73
},
{
"epoch": 0.08491107286288009,
"grad_norm": 0.2360229790210724,
"learning_rate": 4.7588920907110094e-08,
"loss": 0.0118,
"step": 74
},
{
"epoch": 0.08605851979345955,
"grad_norm": 0.244166761636734,
"learning_rate": 0.0,
"loss": 0.0241,
"step": 75
},
{
"epoch": 0.08605851979345955,
"eval_loss": 0.0341278538107872,
"eval_runtime": 54.5534,
"eval_samples_per_second": 13.455,
"eval_steps_per_second": 6.727,
"step": 75
}
],
"logging_steps": 1,
"max_steps": 75,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 25,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.6101318976339968e+16,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}