| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 430, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.3876499831676483, | |
| "learning_rate": 0.00011627906976744187, | |
| "loss": 1.3841, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.35026389360427856, | |
| "learning_rate": 0.0001963824289405685, | |
| "loss": 0.8968, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.317654550075531, | |
| "learning_rate": 0.00018346253229974162, | |
| "loss": 0.6679, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.4258534610271454, | |
| "learning_rate": 0.00017054263565891473, | |
| "loss": 0.5714, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 0.4060456156730652, | |
| "learning_rate": 0.00015762273901808786, | |
| "loss": 0.5099, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 0.37451767921447754, | |
| "learning_rate": 0.000144702842377261, | |
| "loss": 0.5017, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 0.36458975076675415, | |
| "learning_rate": 0.0001317829457364341, | |
| "loss": 0.4647, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "grad_norm": 0.3686697781085968, | |
| "learning_rate": 0.00011886304909560724, | |
| "loss": 0.4359, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "grad_norm": 0.37151384353637695, | |
| "learning_rate": 0.00010594315245478037, | |
| "loss": 0.4268, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "grad_norm": 0.45622408390045166, | |
| "learning_rate": 9.30232558139535e-05, | |
| "loss": 0.4423, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.44779106974601746, | |
| "learning_rate": 8.010335917312663e-05, | |
| "loss": 0.3945, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "grad_norm": 0.4048546552658081, | |
| "learning_rate": 6.718346253229974e-05, | |
| "loss": 0.3765, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "grad_norm": 0.5087065696716309, | |
| "learning_rate": 5.426356589147287e-05, | |
| "loss": 0.3912, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "grad_norm": 0.5312517285346985, | |
| "learning_rate": 4.1343669250646e-05, | |
| "loss": 0.3972, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "grad_norm": 0.47890299558639526, | |
| "learning_rate": 2.842377260981912e-05, | |
| "loss": 0.3485, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "grad_norm": 0.5938854217529297, | |
| "learning_rate": 1.5503875968992248e-05, | |
| "loss": 0.3707, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "grad_norm": 0.5553755760192871, | |
| "learning_rate": 2.583979328165375e-06, | |
| "loss": 0.3336, | |
| "step": 425 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 430, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 6.991294110892032e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |