hr16's picture
Training in progress, epoch 1, checkpoint
25a5bd2 verified
raw
history blame
13.5 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 500,
"global_step": 1469,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.013614703880190605,
"grad_norm": 46457.98828125,
"learning_rate": 8.163265306122449e-07,
"loss": 2.0792,
"step": 20
},
{
"epoch": 0.02722940776038121,
"grad_norm": 35946.40234375,
"learning_rate": 1.6326530612244897e-06,
"loss": 2.0791,
"step": 40
},
{
"epoch": 0.04084411164057182,
"grad_norm": 23938.412109375,
"learning_rate": 2.4489795918367347e-06,
"loss": 2.0782,
"step": 60
},
{
"epoch": 0.05445881552076242,
"grad_norm": 32852.3359375,
"learning_rate": 3.2653061224489794e-06,
"loss": 2.0771,
"step": 80
},
{
"epoch": 0.06807351940095303,
"grad_norm": 32399.712890625,
"learning_rate": 4.081632653061225e-06,
"loss": 2.0752,
"step": 100
},
{
"epoch": 0.08168822328114364,
"grad_norm": 30975.478515625,
"learning_rate": 4.897959183673469e-06,
"loss": 2.0722,
"step": 120
},
{
"epoch": 0.09530292716133425,
"grad_norm": 35666.22265625,
"learning_rate": 5.7142857142857145e-06,
"loss": 2.0674,
"step": 140
},
{
"epoch": 0.10891763104152484,
"grad_norm": 36289.2578125,
"learning_rate": 6.530612244897959e-06,
"loss": 2.0555,
"step": 160
},
{
"epoch": 0.12253233492171545,
"grad_norm": 34320.37890625,
"learning_rate": 7.346938775510204e-06,
"loss": 2.0414,
"step": 180
},
{
"epoch": 0.13614703880190607,
"grad_norm": 49061.66796875,
"learning_rate": 8.16326530612245e-06,
"loss": 2.0297,
"step": 200
},
{
"epoch": 0.14976174268209666,
"grad_norm": 60060.45703125,
"learning_rate": 8.979591836734694e-06,
"loss": 2.0094,
"step": 220
},
{
"epoch": 0.16337644656228728,
"grad_norm": 80072.6015625,
"learning_rate": 9.795918367346939e-06,
"loss": 1.9789,
"step": 240
},
{
"epoch": 0.17699115044247787,
"grad_norm": 95368.8828125,
"learning_rate": 1.0612244897959184e-05,
"loss": 1.9378,
"step": 260
},
{
"epoch": 0.1906058543226685,
"grad_norm": 70028.4765625,
"learning_rate": 1.1428571428571429e-05,
"loss": 1.9166,
"step": 280
},
{
"epoch": 0.2042205582028591,
"grad_norm": 90220.5859375,
"learning_rate": 1.2244897959183674e-05,
"loss": 1.8492,
"step": 300
},
{
"epoch": 0.21783526208304968,
"grad_norm": 59154.35546875,
"learning_rate": 1.3061224489795918e-05,
"loss": 1.8556,
"step": 320
},
{
"epoch": 0.2314499659632403,
"grad_norm": 129406.875,
"learning_rate": 1.3877551020408165e-05,
"loss": 1.8048,
"step": 340
},
{
"epoch": 0.2450646698434309,
"grad_norm": 176714.859375,
"learning_rate": 1.4693877551020408e-05,
"loss": 1.743,
"step": 360
},
{
"epoch": 0.2586793737236215,
"grad_norm": 225369.375,
"learning_rate": 1.5510204081632655e-05,
"loss": 1.708,
"step": 380
},
{
"epoch": 0.27229407760381213,
"grad_norm": 324853.53125,
"learning_rate": 1.63265306122449e-05,
"loss": 1.7032,
"step": 400
},
{
"epoch": 0.2859087814840027,
"grad_norm": 77990.5,
"learning_rate": 1.7142857142857142e-05,
"loss": 1.7964,
"step": 420
},
{
"epoch": 0.2995234853641933,
"grad_norm": 243999.53125,
"learning_rate": 1.7959183673469387e-05,
"loss": 1.7472,
"step": 440
},
{
"epoch": 0.3131381892443839,
"grad_norm": 402316.78125,
"learning_rate": 1.8775510204081636e-05,
"loss": 1.6196,
"step": 460
},
{
"epoch": 0.32675289312457456,
"grad_norm": 173016.109375,
"learning_rate": 1.9591836734693877e-05,
"loss": 1.6543,
"step": 480
},
{
"epoch": 0.34036759700476515,
"grad_norm": 166789.75,
"learning_rate": 2.0408163265306123e-05,
"loss": 1.7865,
"step": 500
},
{
"epoch": 0.34036759700476515,
"eval_accuracy": 0.1981891348088531,
"eval_loss": 1.6593838930130005,
"eval_runtime": 296.9022,
"eval_samples_per_second": 10.044,
"eval_steps_per_second": 1.256,
"step": 500
},
{
"epoch": 0.35398230088495575,
"grad_norm": 160284.609375,
"learning_rate": 2.1224489795918368e-05,
"loss": 1.6528,
"step": 520
},
{
"epoch": 0.36759700476514634,
"grad_norm": 195971.28125,
"learning_rate": 2.2040816326530613e-05,
"loss": 1.7909,
"step": 540
},
{
"epoch": 0.381211708645337,
"grad_norm": 152683.9375,
"learning_rate": 2.2857142857142858e-05,
"loss": 1.6901,
"step": 560
},
{
"epoch": 0.3948264125255276,
"grad_norm": 188117.421875,
"learning_rate": 2.3673469387755103e-05,
"loss": 1.7382,
"step": 580
},
{
"epoch": 0.4084411164057182,
"grad_norm": 221595.171875,
"learning_rate": 2.448979591836735e-05,
"loss": 1.7579,
"step": 600
},
{
"epoch": 0.42205582028590877,
"grad_norm": 250219.484375,
"learning_rate": 2.5306122448979594e-05,
"loss": 1.6842,
"step": 620
},
{
"epoch": 0.43567052416609936,
"grad_norm": 362297.6875,
"learning_rate": 2.6122448979591835e-05,
"loss": 1.6874,
"step": 640
},
{
"epoch": 0.44928522804629,
"grad_norm": 268312.65625,
"learning_rate": 2.6938775510204084e-05,
"loss": 1.6339,
"step": 660
},
{
"epoch": 0.4628999319264806,
"grad_norm": 301583.75,
"learning_rate": 2.775510204081633e-05,
"loss": 1.6268,
"step": 680
},
{
"epoch": 0.4765146358066712,
"grad_norm": 168292.046875,
"learning_rate": 2.857142857142857e-05,
"loss": 1.7985,
"step": 700
},
{
"epoch": 0.4901293396868618,
"grad_norm": 274193.40625,
"learning_rate": 2.9387755102040816e-05,
"loss": 1.7689,
"step": 720
},
{
"epoch": 0.5037440435670524,
"grad_norm": 142488.421875,
"learning_rate": 2.9977307110438728e-05,
"loss": 1.7201,
"step": 740
},
{
"epoch": 0.517358747447243,
"grad_norm": 273465.96875,
"learning_rate": 2.9886535552193645e-05,
"loss": 1.6105,
"step": 760
},
{
"epoch": 0.5309734513274337,
"grad_norm": 408773.3125,
"learning_rate": 2.9795763993948565e-05,
"loss": 1.6843,
"step": 780
},
{
"epoch": 0.5445881552076243,
"grad_norm": 187610.140625,
"learning_rate": 2.970499243570348e-05,
"loss": 1.7259,
"step": 800
},
{
"epoch": 0.5582028590878149,
"grad_norm": 141979.921875,
"learning_rate": 2.9614220877458398e-05,
"loss": 1.7572,
"step": 820
},
{
"epoch": 0.5718175629680055,
"grad_norm": 183132.390625,
"learning_rate": 2.9523449319213314e-05,
"loss": 1.7094,
"step": 840
},
{
"epoch": 0.585432266848196,
"grad_norm": 299271.59375,
"learning_rate": 2.943267776096823e-05,
"loss": 1.7294,
"step": 860
},
{
"epoch": 0.5990469707283866,
"grad_norm": 207611.09375,
"learning_rate": 2.9341906202723147e-05,
"loss": 1.7864,
"step": 880
},
{
"epoch": 0.6126616746085772,
"grad_norm": 262759.0625,
"learning_rate": 2.9251134644478064e-05,
"loss": 1.7288,
"step": 900
},
{
"epoch": 0.6262763784887678,
"grad_norm": 291027.125,
"learning_rate": 2.916036308623298e-05,
"loss": 1.6042,
"step": 920
},
{
"epoch": 0.6398910823689585,
"grad_norm": 355390.71875,
"learning_rate": 2.90695915279879e-05,
"loss": 1.7426,
"step": 940
},
{
"epoch": 0.6535057862491491,
"grad_norm": 87989.2578125,
"learning_rate": 2.8978819969742813e-05,
"loss": 1.6938,
"step": 960
},
{
"epoch": 0.6671204901293397,
"grad_norm": 302982.09375,
"learning_rate": 2.888804841149773e-05,
"loss": 1.8035,
"step": 980
},
{
"epoch": 0.6807351940095303,
"grad_norm": 197021.3125,
"learning_rate": 2.879727685325265e-05,
"loss": 1.7716,
"step": 1000
},
{
"epoch": 0.6807351940095303,
"eval_accuracy": 0.1981891348088531,
"eval_loss": 1.6633707284927368,
"eval_runtime": 291.4938,
"eval_samples_per_second": 10.23,
"eval_steps_per_second": 1.28,
"step": 1000
},
{
"epoch": 0.6943498978897209,
"grad_norm": 348515.0625,
"learning_rate": 2.8706505295007566e-05,
"loss": 1.7593,
"step": 1020
},
{
"epoch": 0.7079646017699115,
"grad_norm": 252458.609375,
"learning_rate": 2.8615733736762483e-05,
"loss": 1.7406,
"step": 1040
},
{
"epoch": 0.7215793056501021,
"grad_norm": 179026.3125,
"learning_rate": 2.8524962178517396e-05,
"loss": 1.806,
"step": 1060
},
{
"epoch": 0.7351940095302927,
"grad_norm": 230475.78125,
"learning_rate": 2.8434190620272316e-05,
"loss": 1.7327,
"step": 1080
},
{
"epoch": 0.7488087134104833,
"grad_norm": 155040.359375,
"learning_rate": 2.8343419062027232e-05,
"loss": 1.6378,
"step": 1100
},
{
"epoch": 0.762423417290674,
"grad_norm": 200565.359375,
"learning_rate": 2.825264750378215e-05,
"loss": 1.7561,
"step": 1120
},
{
"epoch": 0.7760381211708646,
"grad_norm": 160959.6875,
"learning_rate": 2.8161875945537065e-05,
"loss": 1.7322,
"step": 1140
},
{
"epoch": 0.7896528250510552,
"grad_norm": 277028.65625,
"learning_rate": 2.8071104387291985e-05,
"loss": 1.6283,
"step": 1160
},
{
"epoch": 0.8032675289312458,
"grad_norm": 145514.578125,
"learning_rate": 2.79803328290469e-05,
"loss": 1.696,
"step": 1180
},
{
"epoch": 0.8168822328114363,
"grad_norm": 274652.6875,
"learning_rate": 2.7889561270801815e-05,
"loss": 1.7019,
"step": 1200
},
{
"epoch": 0.8304969366916269,
"grad_norm": 108995.4453125,
"learning_rate": 2.779878971255673e-05,
"loss": 1.7189,
"step": 1220
},
{
"epoch": 0.8441116405718175,
"grad_norm": 84566.8046875,
"learning_rate": 2.770801815431165e-05,
"loss": 1.8068,
"step": 1240
},
{
"epoch": 0.8577263444520081,
"grad_norm": 239363.8125,
"learning_rate": 2.7617246596066568e-05,
"loss": 1.8275,
"step": 1260
},
{
"epoch": 0.8713410483321987,
"grad_norm": 213879.875,
"learning_rate": 2.752647503782148e-05,
"loss": 1.7145,
"step": 1280
},
{
"epoch": 0.8849557522123894,
"grad_norm": 201732.859375,
"learning_rate": 2.74357034795764e-05,
"loss": 1.648,
"step": 1300
},
{
"epoch": 0.89857045609258,
"grad_norm": 262597.78125,
"learning_rate": 2.7344931921331318e-05,
"loss": 1.6619,
"step": 1320
},
{
"epoch": 0.9121851599727706,
"grad_norm": 343452.15625,
"learning_rate": 2.7254160363086234e-05,
"loss": 1.6732,
"step": 1340
},
{
"epoch": 0.9257998638529612,
"grad_norm": 164955.390625,
"learning_rate": 2.716338880484115e-05,
"loss": 1.837,
"step": 1360
},
{
"epoch": 0.9394145677331518,
"grad_norm": 243246.953125,
"learning_rate": 2.7072617246596067e-05,
"loss": 1.7863,
"step": 1380
},
{
"epoch": 0.9530292716133424,
"grad_norm": 133524.296875,
"learning_rate": 2.6981845688350984e-05,
"loss": 1.671,
"step": 1400
},
{
"epoch": 0.966643975493533,
"grad_norm": 187500.75,
"learning_rate": 2.68910741301059e-05,
"loss": 1.6882,
"step": 1420
},
{
"epoch": 0.9802586793737236,
"grad_norm": 311251.9375,
"learning_rate": 2.6800302571860817e-05,
"loss": 1.6963,
"step": 1440
},
{
"epoch": 0.9938733832539143,
"grad_norm": 451302.75,
"learning_rate": 2.6709531013615737e-05,
"loss": 1.7184,
"step": 1460
}
],
"logging_steps": 20,
"max_steps": 7345,
"num_input_tokens_seen": 0,
"num_train_epochs": 5,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 3.24301109962752e+18,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}