|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 1.0, |
|
"eval_steps": 30, |
|
"global_step": 450, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.022222222222222223, |
|
"grad_norm": 0.46052348613739014, |
|
"learning_rate": 9.987820251299122e-05, |
|
"loss": 0.3017, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.044444444444444446, |
|
"grad_norm": 0.0518617108464241, |
|
"learning_rate": 9.951340343707852e-05, |
|
"loss": 0.0063, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.06666666666666667, |
|
"grad_norm": 0.02703821286559105, |
|
"learning_rate": 9.890738003669029e-05, |
|
"loss": 0.0041, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.06666666666666667, |
|
"eval_loss": 0.004034009762108326, |
|
"eval_runtime": 104.0521, |
|
"eval_samples_per_second": 9.611, |
|
"eval_steps_per_second": 0.481, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.08888888888888889, |
|
"grad_norm": 0.011427383869886398, |
|
"learning_rate": 9.806308479691595e-05, |
|
"loss": 0.004, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.1111111111111111, |
|
"grad_norm": 0.022027652710676193, |
|
"learning_rate": 9.698463103929542e-05, |
|
"loss": 0.004, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.13333333333333333, |
|
"grad_norm": 0.001629775040782988, |
|
"learning_rate": 9.567727288213005e-05, |
|
"loss": 0.0039, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.13333333333333333, |
|
"eval_loss": 0.0039054970256984234, |
|
"eval_runtime": 104.0544, |
|
"eval_samples_per_second": 9.61, |
|
"eval_steps_per_second": 0.481, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.15555555555555556, |
|
"grad_norm": 0.04784173518419266, |
|
"learning_rate": 9.414737964294636e-05, |
|
"loss": 0.0039, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.17777777777777778, |
|
"grad_norm": 0.020441358909010887, |
|
"learning_rate": 9.24024048078213e-05, |
|
"loss": 0.0039, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"grad_norm": 0.0022530381102114916, |
|
"learning_rate": 9.045084971874738e-05, |
|
"loss": 0.0039, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"eval_loss": 0.0038894277531653643, |
|
"eval_runtime": 169.6744, |
|
"eval_samples_per_second": 5.894, |
|
"eval_steps_per_second": 0.295, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.2222222222222222, |
|
"grad_norm": 0.006415847223252058, |
|
"learning_rate": 8.83022221559489e-05, |
|
"loss": 0.0039, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.24444444444444444, |
|
"grad_norm": 0.0103780971840024, |
|
"learning_rate": 8.596699001693255e-05, |
|
"loss": 0.0039, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.26666666666666666, |
|
"grad_norm": 0.016158604994416237, |
|
"learning_rate": 8.345653031794292e-05, |
|
"loss": 0.0039, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.26666666666666666, |
|
"eval_loss": 0.00390805397182703, |
|
"eval_runtime": 104.1821, |
|
"eval_samples_per_second": 9.599, |
|
"eval_steps_per_second": 0.48, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.28888888888888886, |
|
"grad_norm": 0.004166940692812204, |
|
"learning_rate": 8.07830737662829e-05, |
|
"loss": 0.0039, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.3111111111111111, |
|
"grad_norm": 0.0024945680052042007, |
|
"learning_rate": 7.795964517353735e-05, |
|
"loss": 0.0039, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.3333333333333333, |
|
"grad_norm": 0.024202220141887665, |
|
"learning_rate": 7.500000000000001e-05, |
|
"loss": 0.0038, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.3333333333333333, |
|
"eval_loss": 0.00407786201685667, |
|
"eval_runtime": 104.0841, |
|
"eval_samples_per_second": 9.608, |
|
"eval_steps_per_second": 0.48, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.35555555555555557, |
|
"grad_norm": 0.005438762251287699, |
|
"learning_rate": 7.191855733945387e-05, |
|
"loss": 0.0039, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.37777777777777777, |
|
"grad_norm": 0.008657037280499935, |
|
"learning_rate": 6.873032967079561e-05, |
|
"loss": 0.0039, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"grad_norm": 0.014475060626864433, |
|
"learning_rate": 6.545084971874738e-05, |
|
"loss": 0.0038, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"eval_loss": 0.0037861280143260956, |
|
"eval_runtime": 104.145, |
|
"eval_samples_per_second": 9.602, |
|
"eval_steps_per_second": 0.48, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.4222222222222222, |
|
"grad_norm": 0.013048152439296246, |
|
"learning_rate": 6.209609477998338e-05, |
|
"loss": 0.0037, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 0.4444444444444444, |
|
"grad_norm": 0.04640911519527435, |
|
"learning_rate": 5.868240888334653e-05, |
|
"loss": 0.0034, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.4666666666666667, |
|
"grad_norm": 0.019521236419677734, |
|
"learning_rate": 5.522642316338268e-05, |
|
"loss": 0.003, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 0.4666666666666667, |
|
"eval_loss": 0.002776140347123146, |
|
"eval_runtime": 104.0124, |
|
"eval_samples_per_second": 9.614, |
|
"eval_steps_per_second": 0.481, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 0.4888888888888889, |
|
"grad_norm": 0.03593623265624046, |
|
"learning_rate": 5.174497483512506e-05, |
|
"loss": 0.0031, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.5111111111111111, |
|
"grad_norm": 0.06989370286464691, |
|
"learning_rate": 4.825502516487497e-05, |
|
"loss": 0.0028, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 0.5333333333333333, |
|
"grad_norm": 0.022669225931167603, |
|
"learning_rate": 4.477357683661734e-05, |
|
"loss": 0.0029, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.5333333333333333, |
|
"eval_loss": 0.0026384310331195593, |
|
"eval_runtime": 104.0301, |
|
"eval_samples_per_second": 9.613, |
|
"eval_steps_per_second": 0.481, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.5555555555555556, |
|
"grad_norm": 0.030363447964191437, |
|
"learning_rate": 4.131759111665349e-05, |
|
"loss": 0.0028, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.5777777777777777, |
|
"grad_norm": 0.0183635912835598, |
|
"learning_rate": 3.790390522001662e-05, |
|
"loss": 0.0027, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"grad_norm": 0.04144422709941864, |
|
"learning_rate": 3.4549150281252636e-05, |
|
"loss": 0.0027, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"eval_loss": 0.0025762419681996107, |
|
"eval_runtime": 104.0363, |
|
"eval_samples_per_second": 9.612, |
|
"eval_steps_per_second": 0.481, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 0.6222222222222222, |
|
"grad_norm": 0.030761701986193657, |
|
"learning_rate": 3.12696703292044e-05, |
|
"loss": 0.0027, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.6444444444444445, |
|
"grad_norm": 0.017133990302681923, |
|
"learning_rate": 2.8081442660546125e-05, |
|
"loss": 0.0024, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 0.6666666666666666, |
|
"grad_norm": 0.03892870247364044, |
|
"learning_rate": 2.500000000000001e-05, |
|
"loss": 0.0027, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.6666666666666666, |
|
"eval_loss": 0.002488846192136407, |
|
"eval_runtime": 104.0584, |
|
"eval_samples_per_second": 9.61, |
|
"eval_steps_per_second": 0.48, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.6888888888888889, |
|
"grad_norm": 0.04978968948125839, |
|
"learning_rate": 2.2040354826462668e-05, |
|
"loss": 0.0024, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 0.7111111111111111, |
|
"grad_norm": 0.059906598180532455, |
|
"learning_rate": 1.9216926233717085e-05, |
|
"loss": 0.0029, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 0.7333333333333333, |
|
"grad_norm": 0.015218794345855713, |
|
"learning_rate": 1.6543469682057106e-05, |
|
"loss": 0.0027, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 0.7333333333333333, |
|
"eval_loss": 0.002476006979122758, |
|
"eval_runtime": 103.9799, |
|
"eval_samples_per_second": 9.617, |
|
"eval_steps_per_second": 0.481, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 0.7555555555555555, |
|
"grad_norm": 0.019350852817296982, |
|
"learning_rate": 1.4033009983067452e-05, |
|
"loss": 0.0025, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 0.7777777777777778, |
|
"grad_norm": 0.013380717486143112, |
|
"learning_rate": 1.1697777844051105e-05, |
|
"loss": 0.0028, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"grad_norm": 0.02001432701945305, |
|
"learning_rate": 9.549150281252633e-06, |
|
"loss": 0.0026, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"eval_loss": 0.0024831434711813927, |
|
"eval_runtime": 103.9994, |
|
"eval_samples_per_second": 9.615, |
|
"eval_steps_per_second": 0.481, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 0.8222222222222222, |
|
"grad_norm": 0.015223371796309948, |
|
"learning_rate": 7.597595192178702e-06, |
|
"loss": 0.0028, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 0.8444444444444444, |
|
"grad_norm": 0.025087540969252586, |
|
"learning_rate": 5.852620357053651e-06, |
|
"loss": 0.0023, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 0.8666666666666667, |
|
"grad_norm": 0.02467016503214836, |
|
"learning_rate": 4.322727117869951e-06, |
|
"loss": 0.0026, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 0.8666666666666667, |
|
"eval_loss": 0.0023999777622520924, |
|
"eval_runtime": 104.097, |
|
"eval_samples_per_second": 9.606, |
|
"eval_steps_per_second": 0.48, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 0.8888888888888888, |
|
"grad_norm": 0.043387334793806076, |
|
"learning_rate": 3.0153689607045845e-06, |
|
"loss": 0.0025, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.9111111111111111, |
|
"grad_norm": 0.07373631745576859, |
|
"learning_rate": 1.9369152030840556e-06, |
|
"loss": 0.0028, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 0.9333333333333333, |
|
"grad_norm": 0.05141167715191841, |
|
"learning_rate": 1.0926199633097157e-06, |
|
"loss": 0.0024, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 0.9333333333333333, |
|
"eval_loss": 0.002397245028987527, |
|
"eval_runtime": 104.0308, |
|
"eval_samples_per_second": 9.613, |
|
"eval_steps_per_second": 0.481, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 0.9555555555555556, |
|
"grad_norm": 0.03251807764172554, |
|
"learning_rate": 4.865965629214819e-07, |
|
"loss": 0.0024, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 0.9777777777777777, |
|
"grad_norm": 0.028111129999160767, |
|
"learning_rate": 1.2179748700879012e-07, |
|
"loss": 0.0023, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"grad_norm": 0.022347470745444298, |
|
"learning_rate": 0.0, |
|
"loss": 0.0025, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 0.0023939567618072033, |
|
"eval_runtime": 103.9872, |
|
"eval_samples_per_second": 9.617, |
|
"eval_steps_per_second": 0.481, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"step": 450, |
|
"total_flos": 7.36157876644479e+17, |
|
"train_loss": 0.00987041109551986, |
|
"train_runtime": 4750.9394, |
|
"train_samples_per_second": 1.894, |
|
"train_steps_per_second": 0.095 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 450, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 7.36157876644479e+17, |
|
"train_batch_size": 10, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|