checkpoint-130/trainer_state.json

{
 "best_metric": null,
 "best_model_checkpoint": null,
 "epoch": 0.9961685823754789,
 "eval_steps": 500,
 "global_step": 130,
 "is_hyper_param_search": false,
 "is_local_process_zero": true,
 "is_world_process_zero": true,
 "log_history": [
 {
 "epoch": 0.19,
 "grad_norm": 1.490241289138794,
 "learning_rate": 0.0001811965811965812,
 "loss": 5.1039,
 "step": 25
 },
 {
 "epoch": 0.38,
 "grad_norm": 0.7702738046646118,
 "learning_rate": 0.00013846153846153847,
 "loss": 2.7162,
 "step": 50
 },
 {
 "epoch": 0.57,
 "grad_norm": 2.0000827312469482,
 "learning_rate": 9.572649572649574e-05,
 "loss": 2.4764,
 "step": 75
 },
 {
 "epoch": 0.77,
 "grad_norm": 0.692040205001831,
 "learning_rate": 5.2991452991453e-05,
 "loss": 2.4642,
 "step": 100
 },
 {
 "epoch": 0.96,
 "grad_norm": 0.8987083435058594,
 "learning_rate": 1.0256410256410256e-05,
 "loss": 2.3654,
 "step": 125
 }
 ],
 "logging_steps": 25,
 "max_steps": 130,
 "num_input_tokens_seen": 0,
 "num_train_epochs": 1,
 "save_steps": 500,
 "total_flos": 2.11364836540416e+16,
 "train_batch_size": 1,
 "trial_name": null,
 "trial_params": null
}