trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.24582104228121926,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "grad_norm": 0.362138569355011,
      "learning_rate": 0.002,
      "loss": 1.871,
      "step": 100
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.22625844180583954,
      "learning_rate": 0.002,
      "loss": 1.8426,
      "step": 200
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.2156531810760498,
      "learning_rate": 0.002,
      "loss": 1.822,
      "step": 300
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.23074010014533997,
      "learning_rate": 0.002,
      "loss": 1.8194,
      "step": 400
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.2703346908092499,
      "learning_rate": 0.002,
      "loss": 1.8077,
      "step": 500
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.2685506343841553,
      "learning_rate": 0.002,
      "loss": 1.7923,
      "step": 600
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.279742956161499,
      "learning_rate": 0.002,
      "loss": 1.7939,
      "step": 700
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.2931801974773407,
      "learning_rate": 0.002,
      "loss": 1.7901,
      "step": 800
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.3071358799934387,
      "learning_rate": 0.002,
      "loss": 1.7811,
      "step": 900
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.24978914856910706,
      "learning_rate": 0.002,
      "loss": 1.7598,
      "step": 1000
    }
  ],
  "logging_steps": 100,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 8242039250583552.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}