trainer_state.json

{
 "best_metric": null,
 "best_model_checkpoint": null,
 "epoch": 1.0,
 "eval_steps": 500,
 "global_step": 253,
 "is_hyper_param_search": false,
 "is_local_process_zero": true,
 "is_world_process_zero": true,
 "log_history": [
 {
 "epoch": 0.0,
 "learning_rate": 2e-05,
 "loss": 2.5574,
 "step": 1
 },
 {
 "epoch": 1.0,
 "eval_loss": 1.9306640625,
 "eval_runtime": 4.829,
 "eval_samples_per_second": 2956.915,
 "eval_steps_per_second": 5.798,
 "step": 253
 },
 {
 "epoch": 1.0,
 "step": 253,
 "total_flos": 180169583099904.0,
 "train_loss": 2.0571069528934043,
 "train_runtime": 132.1266,
 "train_samples_per_second": 977.055,
 "train_steps_per_second": 1.915
 }
 ],
 "logging_steps": 500,
 "max_steps": 253,
 "num_input_tokens_seen": 0,
 "num_train_epochs": 1,
 "save_steps": 500,
 "total_flos": 180169583099904.0,
 "train_batch_size": 16,
 "trial_name": null,
 "trial_params": null
}