{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.028276544606249118,
"eval_steps": 500,
"global_step": 50,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0005655308921249824,
"grad_norm": 9.447598457336426,
"learning_rate": 0.0001,
"loss": 7.5877,
"step": 1
},
{
"epoch": 0.0011310617842499647,
"grad_norm": 12.224023818969727,
"learning_rate": 0.0002,
"loss": 7.4966,
"step": 2
},
{
"epoch": 0.001696592676374947,
"grad_norm": 8.18627643585205,
"learning_rate": 0.00019978589232386035,
"loss": 6.524,
"step": 3
},
{
"epoch": 0.0022621235684999294,
"grad_norm": 9.624427795410156,
"learning_rate": 0.00019914448613738106,
"loss": 4.5356,
"step": 4
},
{
"epoch": 0.0028276544606249117,
"grad_norm": 8.553901672363281,
"learning_rate": 0.00019807852804032305,
"loss": 4.2528,
"step": 5
},
{
"epoch": 0.003393185352749894,
"grad_norm": 5.756651401519775,
"learning_rate": 0.00019659258262890683,
"loss": 4.1727,
"step": 6
},
{
"epoch": 0.003958716244874877,
"grad_norm": 2.862290143966675,
"learning_rate": 0.0001946930129495106,
"loss": 4.1679,
"step": 7
},
{
"epoch": 0.004524247136999859,
"grad_norm": 1.6996726989746094,
"learning_rate": 0.0001923879532511287,
"loss": 4.1149,
"step": 8
},
{
"epoch": 0.005089778029124841,
"grad_norm": 2.1115152835845947,
"learning_rate": 0.00018968727415326884,
"loss": 3.8839,
"step": 9
},
{
"epoch": 0.005655308921249823,
"grad_norm": 1.4950510263442993,
"learning_rate": 0.00018660254037844388,
"loss": 4.0708,
"step": 10
},
{
"epoch": 0.006220839813374806,
"grad_norm": 3.2026827335357666,
"learning_rate": 0.00018314696123025454,
"loss": 4.2101,
"step": 11
},
{
"epoch": 0.006786370705499788,
"grad_norm": 1.4218138456344604,
"learning_rate": 0.00017933533402912354,
"loss": 4.0784,
"step": 12
},
{
"epoch": 0.00735190159762477,
"grad_norm": 2.44765043258667,
"learning_rate": 0.00017518398074789775,
"loss": 4.079,
"step": 13
},
{
"epoch": 0.007917432489749753,
"grad_norm": 1.5132642984390259,
"learning_rate": 0.00017071067811865476,
"loss": 4.1191,
"step": 14
},
{
"epoch": 0.008482963381874735,
"grad_norm": 1.7532415390014648,
"learning_rate": 0.00016593458151000688,
"loss": 3.9846,
"step": 15
},
{
"epoch": 0.009048494273999718,
"grad_norm": 1.528114914894104,
"learning_rate": 0.00016087614290087208,
"loss": 4.0772,
"step": 16
},
{
"epoch": 0.009614025166124699,
"grad_norm": 1.0905526876449585,
"learning_rate": 0.00015555702330196023,
"loss": 4.1484,
"step": 17
},
{
"epoch": 0.010179556058249682,
"grad_norm": 1.0690267086029053,
"learning_rate": 0.00015000000000000001,
"loss": 4.01,
"step": 18
},
{
"epoch": 0.010745086950374664,
"grad_norm": 0.7005692720413208,
"learning_rate": 0.00014422886902190014,
"loss": 3.9638,
"step": 19
},
{
"epoch": 0.011310617842499647,
"grad_norm": 0.8180456757545471,
"learning_rate": 0.000138268343236509,
"loss": 4.0638,
"step": 20
},
{
"epoch": 0.011876148734624628,
"grad_norm": 0.7685333490371704,
"learning_rate": 0.00013214394653031616,
"loss": 4.0947,
"step": 21
},
{
"epoch": 0.012441679626749611,
"grad_norm": 0.8267884850502014,
"learning_rate": 0.00012588190451025207,
"loss": 4.1512,
"step": 22
},
{
"epoch": 0.013007210518874594,
"grad_norm": 0.80919349193573,
"learning_rate": 0.00011950903220161285,
"loss": 4.0781,
"step": 23
},
{
"epoch": 0.013572741410999576,
"grad_norm": 0.862808108329773,
"learning_rate": 0.00011305261922200519,
"loss": 4.0492,
"step": 24
},
{
"epoch": 0.014138272303124559,
"grad_norm": 1.0544193983078003,
"learning_rate": 0.00010654031292301432,
"loss": 3.9988,
"step": 25
},
{
"epoch": 0.01470380319524954,
"grad_norm": 0.8450819253921509,
"learning_rate": 0.0001,
"loss": 4.0739,
"step": 26
},
{
"epoch": 0.015269334087374523,
"grad_norm": 1.9969067573547363,
"learning_rate": 9.345968707698569e-05,
"loss": 3.9674,
"step": 27
},
{
"epoch": 0.015834864979499506,
"grad_norm": 0.7638323903083801,
"learning_rate": 8.694738077799488e-05,
"loss": 3.9486,
"step": 28
},
{
"epoch": 0.016400395871624486,
"grad_norm": 0.6288601756095886,
"learning_rate": 8.049096779838719e-05,
"loss": 3.8704,
"step": 29
},
{
"epoch": 0.01696592676374947,
"grad_norm": 1.469429612159729,
"learning_rate": 7.411809548974792e-05,
"loss": 4.1891,
"step": 30
},
{
"epoch": 0.017531457655874452,
"grad_norm": 1.1576073169708252,
"learning_rate": 6.785605346968386e-05,
"loss": 3.986,
"step": 31
},
{
"epoch": 0.018096988547999435,
"grad_norm": 0.9627673029899597,
"learning_rate": 6.173165676349103e-05,
"loss": 4.1144,
"step": 32
},
{
"epoch": 0.01866251944012442,
"grad_norm": 0.7569557428359985,
"learning_rate": 5.577113097809989e-05,
"loss": 4.0838,
"step": 33
},
{
"epoch": 0.019228050332249398,
"grad_norm": 0.7780810594558716,
"learning_rate": 5.000000000000002e-05,
"loss": 4.0803,
"step": 34
},
{
"epoch": 0.01979358122437438,
"grad_norm": 1.5027967691421509,
"learning_rate": 4.444297669803981e-05,
"loss": 3.9828,
"step": 35
},
{
"epoch": 0.020359112116499364,
"grad_norm": 0.9394671320915222,
"learning_rate": 3.9123857099127936e-05,
"loss": 4.0787,
"step": 36
},
{
"epoch": 0.020924643008624347,
"grad_norm": 0.8779884576797485,
"learning_rate": 3.406541848999312e-05,
"loss": 4.0903,
"step": 37
},
{
"epoch": 0.021490173900749327,
"grad_norm": 1.4611361026763916,
"learning_rate": 2.9289321881345254e-05,
"loss": 3.9183,
"step": 38
},
{
"epoch": 0.02205570479287431,
"grad_norm": 0.7395803928375244,
"learning_rate": 2.4816019252102273e-05,
"loss": 4.1387,
"step": 39
},
{
"epoch": 0.022621235684999293,
"grad_norm": 0.8143374919891357,
"learning_rate": 2.0664665970876496e-05,
"loss": 4.1567,
"step": 40
},
{
"epoch": 0.023186766577124276,
"grad_norm": 0.7555928826332092,
"learning_rate": 1.6853038769745467e-05,
"loss": 4.0109,
"step": 41
},
{
"epoch": 0.023752297469249256,
"grad_norm": 0.749727189540863,
"learning_rate": 1.339745962155613e-05,
"loss": 4.0942,
"step": 42
},
{
"epoch": 0.02431782836137424,
"grad_norm": 2.71633243560791,
"learning_rate": 1.0312725846731175e-05,
"loss": 3.978,
"step": 43
},
{
"epoch": 0.024883359253499222,
"grad_norm": 0.856410026550293,
"learning_rate": 7.612046748871327e-06,
"loss": 3.8863,
"step": 44
},
{
"epoch": 0.025448890145624205,
"grad_norm": 0.9397003054618835,
"learning_rate": 5.306987050489442e-06,
"loss": 3.8758,
"step": 45
},
{
"epoch": 0.02601442103774919,
"grad_norm": 0.8372579216957092,
"learning_rate": 3.40741737109318e-06,
"loss": 4.1118,
"step": 46
},
{
"epoch": 0.026579951929874168,
"grad_norm": 0.6836534142494202,
"learning_rate": 1.921471959676957e-06,
"loss": 4.0193,
"step": 47
},
{
"epoch": 0.02714548282199915,
"grad_norm": 0.7352842688560486,
"learning_rate": 8.555138626189618e-07,
"loss": 3.9597,
"step": 48
},
{
"epoch": 0.027711013714124134,
"grad_norm": 1.5941392183303833,
"learning_rate": 2.141076761396521e-07,
"loss": 4.1411,
"step": 49
},
{
"epoch": 0.028276544606249118,
"grad_norm": 1.6705574989318848,
"learning_rate": 0.0,
"loss": 4.1692,
"step": 50
}
],
"logging_steps": 1,
"max_steps": 50,
"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": 8.23612542025728e+16,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}