lesso17's picture
Training in progress, step 500, checkpoint
62449d0 verified
{
"best_metric": 0.832525908946991,
"best_model_checkpoint": "miner_id_24/checkpoint-500",
"epoch": 0.23889154323936931,
"eval_steps": 50,
"global_step": 500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.00047778308647873863,
"eval_loss": 4.254161834716797,
"eval_runtime": 11.5108,
"eval_samples_per_second": 76.624,
"eval_steps_per_second": 19.199,
"step": 1
},
{
"epoch": 0.004777830864787387,
"grad_norm": 86.97021484375,
"learning_rate": 4.34e-05,
"loss": 6.7643,
"step": 10
},
{
"epoch": 0.009555661729574774,
"grad_norm": 114.20069885253906,
"learning_rate": 8.68e-05,
"loss": 4.2519,
"step": 20
},
{
"epoch": 0.01433349259436216,
"grad_norm": 87.37146759033203,
"learning_rate": 0.0001302,
"loss": 3.5983,
"step": 30
},
{
"epoch": 0.019111323459149548,
"grad_norm": 114.74700927734375,
"learning_rate": 0.0001736,
"loss": 3.2439,
"step": 40
},
{
"epoch": 0.023889154323936932,
"grad_norm": 97.87152099609375,
"learning_rate": 0.000217,
"loss": 2.683,
"step": 50
},
{
"epoch": 0.023889154323936932,
"eval_loss": 1.3938980102539062,
"eval_runtime": 11.5211,
"eval_samples_per_second": 76.555,
"eval_steps_per_second": 19.182,
"step": 50
},
{
"epoch": 0.02866698518872432,
"grad_norm": 68.76221466064453,
"learning_rate": 0.00021673569945319091,
"loss": 2.163,
"step": 60
},
{
"epoch": 0.033444816053511704,
"grad_norm": 85.8083267211914,
"learning_rate": 0.00021594408545846038,
"loss": 1.941,
"step": 70
},
{
"epoch": 0.038222646918299095,
"grad_norm": 49.4107780456543,
"learning_rate": 0.0002146290146796179,
"loss": 2.4934,
"step": 80
},
{
"epoch": 0.04300047778308648,
"grad_norm": 45.38887405395508,
"learning_rate": 0.0002127968940093076,
"loss": 2.7001,
"step": 90
},
{
"epoch": 0.047778308647873864,
"grad_norm": 114.47595977783203,
"learning_rate": 0.00021045664935527106,
"loss": 2.8237,
"step": 100
},
{
"epoch": 0.047778308647873864,
"eval_loss": 1.3519835472106934,
"eval_runtime": 11.475,
"eval_samples_per_second": 76.863,
"eval_steps_per_second": 19.259,
"step": 100
},
{
"epoch": 0.05255613951266125,
"grad_norm": 163.51150512695312,
"learning_rate": 0.00020761968215422217,
"loss": 2.9201,
"step": 110
},
{
"epoch": 0.05733397037744864,
"grad_norm": 42.164276123046875,
"learning_rate": 0.00020429981382519356,
"loss": 2.8884,
"step": 120
},
{
"epoch": 0.062111801242236024,
"grad_norm": 30.880720138549805,
"learning_rate": 0.00020051321843297219,
"loss": 2.5363,
"step": 130
},
{
"epoch": 0.06688963210702341,
"grad_norm": 61.331912994384766,
"learning_rate": 0.0001962783438896818,
"loss": 2.4063,
"step": 140
},
{
"epoch": 0.0716674629718108,
"grad_norm": 70.56568145751953,
"learning_rate": 0.0001916158220784091,
"loss": 2.5402,
"step": 150
},
{
"epoch": 0.0716674629718108,
"eval_loss": 1.248333215713501,
"eval_runtime": 11.4863,
"eval_samples_per_second": 76.787,
"eval_steps_per_second": 19.24,
"step": 150
},
{
"epoch": 0.07644529383659819,
"grad_norm": 33.299007415771484,
"learning_rate": 0.00018654836833674362,
"loss": 1.8457,
"step": 160
},
{
"epoch": 0.08122312470138557,
"grad_norm": 48.84244918823242,
"learning_rate": 0.0001811006707899361,
"loss": 2.259,
"step": 170
},
{
"epoch": 0.08600095556617296,
"grad_norm": 16.119049072265625,
"learning_rate": 0.0001752992700728339,
"loss": 2.4819,
"step": 180
},
{
"epoch": 0.09077878643096035,
"grad_norm": 32.0130500793457,
"learning_rate": 0.00016917243002657602,
"loss": 2.4841,
"step": 190
},
{
"epoch": 0.09555661729574773,
"grad_norm": 70.85590362548828,
"learning_rate": 0.00016275,
"loss": 2.8376,
"step": 200
},
{
"epoch": 0.09555661729574773,
"eval_loss": 1.1818195581436157,
"eval_runtime": 11.5022,
"eval_samples_per_second": 76.681,
"eval_steps_per_second": 19.214,
"step": 200
},
{
"epoch": 0.10033444816053512,
"grad_norm": 50.72034454345703,
"learning_rate": 0.0001560632694266149,
"loss": 1.869,
"step": 210
},
{
"epoch": 0.1051122790253225,
"grad_norm": 42.09200668334961,
"learning_rate": 0.00014914481538562646,
"loss": 1.7881,
"step": 220
},
{
"epoch": 0.10989010989010989,
"grad_norm": 118.14753723144531,
"learning_rate": 0.0001420283438896818,
"loss": 2.0519,
"step": 230
},
{
"epoch": 0.11466794075489728,
"grad_norm": 24.315568923950195,
"learning_rate": 0.00013474852567256393,
"loss": 2.1679,
"step": 240
},
{
"epoch": 0.11944577161968466,
"grad_norm": 70.42790985107422,
"learning_rate": 0.00012734082727686196,
"loss": 2.4192,
"step": 250
},
{
"epoch": 0.11944577161968466,
"eval_loss": 1.1315124034881592,
"eval_runtime": 11.4436,
"eval_samples_per_second": 77.074,
"eval_steps_per_second": 19.312,
"step": 250
},
{
"epoch": 0.12422360248447205,
"grad_norm": 41.314945220947266,
"learning_rate": 0.0001198413382645404,
"loss": 1.7405,
"step": 260
},
{
"epoch": 0.12900143334925943,
"grad_norm": 21.883790969848633,
"learning_rate": 0.00011228659539222137,
"loss": 1.6183,
"step": 270
},
{
"epoch": 0.13377926421404682,
"grad_norm": 14.944008827209473,
"learning_rate": 0.00010471340460777866,
"loss": 1.7443,
"step": 280
},
{
"epoch": 0.1385570950788342,
"grad_norm": 38.64680099487305,
"learning_rate": 9.715866173545961e-05,
"loss": 2.1661,
"step": 290
},
{
"epoch": 0.1433349259436216,
"grad_norm": 34.455081939697266,
"learning_rate": 8.965917272313806e-05,
"loss": 2.2251,
"step": 300
},
{
"epoch": 0.1433349259436216,
"eval_loss": 0.9748860597610474,
"eval_runtime": 11.386,
"eval_samples_per_second": 77.464,
"eval_steps_per_second": 19.41,
"step": 300
},
{
"epoch": 0.148112756808409,
"grad_norm": 20.522775650024414,
"learning_rate": 8.225147432743606e-05,
"loss": 1.4992,
"step": 310
},
{
"epoch": 0.15289058767319638,
"grad_norm": 75.11093139648438,
"learning_rate": 7.497165611031821e-05,
"loss": 1.4972,
"step": 320
},
{
"epoch": 0.15766841853798375,
"grad_norm": 37.47844696044922,
"learning_rate": 6.785518461437353e-05,
"loss": 1.8537,
"step": 330
},
{
"epoch": 0.16244624940277114,
"grad_norm": 23.639490127563477,
"learning_rate": 6.093673057338509e-05,
"loss": 2.1694,
"step": 340
},
{
"epoch": 0.16722408026755853,
"grad_norm": 54.70100021362305,
"learning_rate": 5.4250000000000024e-05,
"loss": 2.3268,
"step": 350
},
{
"epoch": 0.16722408026755853,
"eval_loss": 0.9435171484947205,
"eval_runtime": 11.5124,
"eval_samples_per_second": 76.613,
"eval_steps_per_second": 19.197,
"step": 350
},
{
"epoch": 0.17200191113234592,
"grad_norm": 34.21720886230469,
"learning_rate": 4.782756997342398e-05,
"loss": 1.3658,
"step": 360
},
{
"epoch": 0.1767797419971333,
"grad_norm": 22.63213348388672,
"learning_rate": 4.170072992716607e-05,
"loss": 1.4596,
"step": 370
},
{
"epoch": 0.1815575728619207,
"grad_norm": 13.130209922790527,
"learning_rate": 3.5899329210063916e-05,
"loss": 1.6314,
"step": 380
},
{
"epoch": 0.18633540372670807,
"grad_norm": 13.130266189575195,
"learning_rate": 3.045163166325637e-05,
"loss": 1.9832,
"step": 390
},
{
"epoch": 0.19111323459149546,
"grad_norm": 28.62923812866211,
"learning_rate": 2.5384177921590895e-05,
"loss": 2.1456,
"step": 400
},
{
"epoch": 0.19111323459149546,
"eval_loss": 0.8851467370986938,
"eval_runtime": 11.5188,
"eval_samples_per_second": 76.57,
"eval_steps_per_second": 19.186,
"step": 400
},
{
"epoch": 0.19589106545628285,
"grad_norm": 12.487257957458496,
"learning_rate": 2.0721656110318213e-05,
"loss": 1.1234,
"step": 410
},
{
"epoch": 0.20066889632107024,
"grad_norm": 12.503289222717285,
"learning_rate": 1.6486781567027783e-05,
"loss": 1.2566,
"step": 420
},
{
"epoch": 0.20544672718585763,
"grad_norm": 9.571052551269531,
"learning_rate": 1.2700186174806422e-05,
"loss": 1.6166,
"step": 430
},
{
"epoch": 0.210224558050645,
"grad_norm": 11.375801086425781,
"learning_rate": 9.380317845777794e-06,
"loss": 1.9494,
"step": 440
},
{
"epoch": 0.21500238891543239,
"grad_norm": 11.005843162536621,
"learning_rate": 6.543350644728947e-06,
"loss": 2.0351,
"step": 450
},
{
"epoch": 0.21500238891543239,
"eval_loss": 0.8364555239677429,
"eval_runtime": 11.4227,
"eval_samples_per_second": 77.215,
"eval_steps_per_second": 19.347,
"step": 450
},
{
"epoch": 0.21978021978021978,
"grad_norm": 26.848785400390625,
"learning_rate": 4.2031059906924e-06,
"loss": 1.0386,
"step": 460
},
{
"epoch": 0.22455805064500717,
"grad_norm": 13.840141296386719,
"learning_rate": 2.3709853203820825e-06,
"loss": 1.2618,
"step": 470
},
{
"epoch": 0.22933588150979456,
"grad_norm": 37.794471740722656,
"learning_rate": 1.0559145415396157e-06,
"loss": 1.6539,
"step": 480
},
{
"epoch": 0.23411371237458195,
"grad_norm": 38.33562088012695,
"learning_rate": 2.643005468090745e-07,
"loss": 1.7703,
"step": 490
},
{
"epoch": 0.23889154323936931,
"grad_norm": 19.850322723388672,
"learning_rate": 0.0,
"loss": 2.0542,
"step": 500
},
{
"epoch": 0.23889154323936931,
"eval_loss": 0.832525908946991,
"eval_runtime": 11.6134,
"eval_samples_per_second": 75.947,
"eval_steps_per_second": 19.03,
"step": 500
}
],
"logging_steps": 10,
"max_steps": 500,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 50,
"stateful_callbacks": {
"EarlyStoppingCallback": {
"args": {
"early_stopping_patience": 3,
"early_stopping_threshold": 0.0
},
"attributes": {
"early_stopping_patience_counter": 0
}
},
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 4657011228672000.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}