trainer_state.json

{
 "best_metric": null,
 "best_model_checkpoint": null,
 "epoch": 8.0,
 "eval_steps": 500,
 "global_step": 1214,
 "is_hyper_param_search": false,
 "is_local_process_zero": true,
 "is_world_process_zero": true,
 "log_history": [
 {
 "epoch": 0.66,
 "grad_norm": 0.00020524964202195406,
 "learning_rate": 9.364238410596028e-06,
 "loss": 0.0293,
 "step": 100
 },
 {
 "epoch": 1.32,
 "grad_norm": 0.00010464163642609492,
 "learning_rate": 8.701986754966887e-06,
 "loss": 0.0514,
 "step": 200
 },
 {
 "epoch": 1.98,
 "grad_norm": 0.002206347417086363,
 "learning_rate": 8.039735099337748e-06,
 "loss": 0.0168,
 "step": 300
 },
 {
 "epoch": 2.64,
 "grad_norm": 0.005506620742380619,
 "learning_rate": 7.377483443708609e-06,
 "loss": 0.0153,
 "step": 400
 },
 {
 "epoch": 3.29,
 "grad_norm": 0.004578682128340006,
 "learning_rate": 6.7152317880794705e-06,
 "loss": 0.0404,
 "step": 500
 },
 {
 "epoch": 3.95,
 "grad_norm": 0.006431525107473135,
 "learning_rate": 6.052980132450331e-06,
 "loss": 0.0305,
 "step": 600
 },
 {
 "epoch": 4.61,
 "grad_norm": 0.012976592406630516,
 "learning_rate": 5.390728476821192e-06,
 "loss": 0.0099,
 "step": 700
 },
 {
 "epoch": 5.27,
 "grad_norm": 90.73287963867188,
 "learning_rate": 4.728476821192053e-06,
 "loss": 0.0314,
 "step": 800
 },
 {
 "epoch": 5.93,
 "grad_norm": 0.00018791892216540873,
 "learning_rate": 4.066225165562914e-06,
 "loss": 0.023,
 "step": 900
 },
 {
 "epoch": 6.59,
 "grad_norm": 0.0467652902007103,
 "learning_rate": 3.403973509933775e-06,
 "loss": 0.0425,
 "step": 1000
 },
 {
 "epoch": 7.25,
 "grad_norm": 0.0006702475366182625,
 "learning_rate": 2.741721854304636e-06,
 "loss": 0.0139,
 "step": 1100
 },
 {
 "epoch": 7.91,
 "grad_norm": 0.0008735745213925838,
 "learning_rate": 2.079470198675497e-06,
 "loss": 0.0178,
 "step": 1200
 }
 ],
 "logging_steps": 100,
 "max_steps": 1510,
 "num_input_tokens_seen": 0,
 "num_train_epochs": 10,
 "save_steps": 500,
 "total_flos": 0.0,
 "train_batch_size": 1,
 "trial_name": null,
 "trial_params": null
}