Grogros's picture
Training in progress, step 500, checkpoint
0867b04 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.25,
"eval_steps": 500,
"global_step": 500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.005,
"grad_norm": 37.0,
"learning_rate": 1.0000000000000002e-06,
"loss": 1.5359,
"step": 10
},
{
"epoch": 0.01,
"grad_norm": 45.5,
"learning_rate": 2.0000000000000003e-06,
"loss": 1.4797,
"step": 20
},
{
"epoch": 0.015,
"grad_norm": 21.625,
"learning_rate": 3e-06,
"loss": 1.3793,
"step": 30
},
{
"epoch": 0.02,
"grad_norm": 19.625,
"learning_rate": 4.000000000000001e-06,
"loss": 1.26,
"step": 40
},
{
"epoch": 0.025,
"grad_norm": 12.375,
"learning_rate": 5e-06,
"loss": 1.2473,
"step": 50
},
{
"epoch": 0.03,
"grad_norm": 9.75,
"learning_rate": 6e-06,
"loss": 1.132,
"step": 60
},
{
"epoch": 0.035,
"grad_norm": 8.4375,
"learning_rate": 7e-06,
"loss": 1.1854,
"step": 70
},
{
"epoch": 0.04,
"grad_norm": 9.25,
"learning_rate": 8.000000000000001e-06,
"loss": 1.1473,
"step": 80
},
{
"epoch": 0.045,
"grad_norm": 5.5,
"learning_rate": 9e-06,
"loss": 1.1484,
"step": 90
},
{
"epoch": 0.05,
"grad_norm": 4.71875,
"learning_rate": 1e-05,
"loss": 1.1352,
"step": 100
},
{
"epoch": 0.055,
"grad_norm": 5.6875,
"learning_rate": 1.1000000000000001e-05,
"loss": 1.0916,
"step": 110
},
{
"epoch": 0.06,
"grad_norm": 4.90625,
"learning_rate": 1.2e-05,
"loss": 1.0473,
"step": 120
},
{
"epoch": 0.065,
"grad_norm": 5.09375,
"learning_rate": 1.3000000000000001e-05,
"loss": 1.0576,
"step": 130
},
{
"epoch": 0.07,
"grad_norm": 5.6875,
"learning_rate": 1.4e-05,
"loss": 1.0014,
"step": 140
},
{
"epoch": 0.075,
"grad_norm": 5.4375,
"learning_rate": 1.5000000000000002e-05,
"loss": 1.015,
"step": 150
},
{
"epoch": 0.08,
"grad_norm": 5.0625,
"learning_rate": 1.6000000000000003e-05,
"loss": 0.9674,
"step": 160
},
{
"epoch": 0.085,
"grad_norm": 5.125,
"learning_rate": 1.7e-05,
"loss": 0.9754,
"step": 170
},
{
"epoch": 0.09,
"grad_norm": 4.09375,
"learning_rate": 1.8e-05,
"loss": 0.9309,
"step": 180
},
{
"epoch": 0.095,
"grad_norm": 4.875,
"learning_rate": 1.9e-05,
"loss": 0.9158,
"step": 190
},
{
"epoch": 0.1,
"grad_norm": 117.5,
"learning_rate": 2e-05,
"loss": 0.9994,
"step": 200
},
{
"epoch": 0.105,
"grad_norm": 5.34375,
"learning_rate": 1.9998476951563914e-05,
"loss": 1.0623,
"step": 210
},
{
"epoch": 0.11,
"grad_norm": 19.0,
"learning_rate": 1.999390827019096e-05,
"loss": 1.007,
"step": 220
},
{
"epoch": 0.115,
"grad_norm": 5.21875,
"learning_rate": 1.9986295347545738e-05,
"loss": 0.867,
"step": 230
},
{
"epoch": 0.12,
"grad_norm": 3.640625,
"learning_rate": 1.9975640502598243e-05,
"loss": 0.852,
"step": 240
},
{
"epoch": 0.125,
"grad_norm": 4.40625,
"learning_rate": 1.9961946980917457e-05,
"loss": 0.8018,
"step": 250
},
{
"epoch": 0.13,
"grad_norm": 5.125,
"learning_rate": 1.9945218953682736e-05,
"loss": 0.9186,
"step": 260
},
{
"epoch": 0.135,
"grad_norm": 4.875,
"learning_rate": 1.9925461516413224e-05,
"loss": 0.804,
"step": 270
},
{
"epoch": 0.14,
"grad_norm": 3.71875,
"learning_rate": 1.9902680687415704e-05,
"loss": 0.817,
"step": 280
},
{
"epoch": 0.145,
"grad_norm": 4.125,
"learning_rate": 1.9876883405951378e-05,
"loss": 0.7372,
"step": 290
},
{
"epoch": 0.15,
"grad_norm": 3.5,
"learning_rate": 1.9848077530122083e-05,
"loss": 0.7237,
"step": 300
},
{
"epoch": 0.155,
"grad_norm": 3.375,
"learning_rate": 1.9816271834476642e-05,
"loss": 0.6707,
"step": 310
},
{
"epoch": 0.16,
"grad_norm": 4.21875,
"learning_rate": 1.9781476007338058e-05,
"loss": 0.7664,
"step": 320
},
{
"epoch": 0.165,
"grad_norm": 3.640625,
"learning_rate": 1.9743700647852356e-05,
"loss": 0.6838,
"step": 330
},
{
"epoch": 0.17,
"grad_norm": 4.25,
"learning_rate": 1.9702957262759964e-05,
"loss": 0.7225,
"step": 340
},
{
"epoch": 0.175,
"grad_norm": 3.625,
"learning_rate": 1.9659258262890683e-05,
"loss": 0.6754,
"step": 350
},
{
"epoch": 0.18,
"grad_norm": 3.546875,
"learning_rate": 1.961261695938319e-05,
"loss": 0.6344,
"step": 360
},
{
"epoch": 0.185,
"grad_norm": 3.546875,
"learning_rate": 1.9563047559630356e-05,
"loss": 0.6728,
"step": 370
},
{
"epoch": 0.19,
"grad_norm": 4.34375,
"learning_rate": 1.9510565162951538e-05,
"loss": 0.7283,
"step": 380
},
{
"epoch": 0.195,
"grad_norm": 3.671875,
"learning_rate": 1.945518575599317e-05,
"loss": 0.666,
"step": 390
},
{
"epoch": 0.2,
"grad_norm": 4.4375,
"learning_rate": 1.9396926207859085e-05,
"loss": 0.6756,
"step": 400
},
{
"epoch": 0.205,
"grad_norm": 3.71875,
"learning_rate": 1.9335804264972018e-05,
"loss": 0.6558,
"step": 410
},
{
"epoch": 0.21,
"grad_norm": 3.234375,
"learning_rate": 1.9271838545667876e-05,
"loss": 0.6243,
"step": 420
},
{
"epoch": 0.215,
"grad_norm": 3.90625,
"learning_rate": 1.9205048534524405e-05,
"loss": 0.6519,
"step": 430
},
{
"epoch": 0.22,
"grad_norm": 3.375,
"learning_rate": 1.913545457642601e-05,
"loss": 0.626,
"step": 440
},
{
"epoch": 0.225,
"grad_norm": 3.78125,
"learning_rate": 1.9063077870366504e-05,
"loss": 0.6592,
"step": 450
},
{
"epoch": 0.23,
"grad_norm": 3.75,
"learning_rate": 1.8987940462991673e-05,
"loss": 0.5869,
"step": 460
},
{
"epoch": 0.235,
"grad_norm": 4.25,
"learning_rate": 1.891006524188368e-05,
"loss": 0.6157,
"step": 470
},
{
"epoch": 0.24,
"grad_norm": 2.78125,
"learning_rate": 1.8829475928589272e-05,
"loss": 0.5537,
"step": 480
},
{
"epoch": 0.245,
"grad_norm": 2.734375,
"learning_rate": 1.874619707139396e-05,
"loss": 0.631,
"step": 490
},
{
"epoch": 0.25,
"grad_norm": 3.875,
"learning_rate": 1.866025403784439e-05,
"loss": 0.5853,
"step": 500
}
],
"logging_steps": 10,
"max_steps": 2000,
"num_input_tokens_seen": 0,
"num_train_epochs": 9223372036854775807,
"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": 4.7832077697024e+16,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}