trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00041946308724832214,
  "eval_steps": 200,
  "global_step": 4,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00010486577181208053,
      "grad_norm": 96.5,
      "learning_rate": 1.680672268907563e-07,
      "loss": 7.9808,
      "num_input_tokens_seen": 1048576,
      "step": 1
    },
    {
      "epoch": 0.00020973154362416107,
      "grad_norm": 99.0,
      "learning_rate": 3.361344537815126e-07,
      "loss": 8.0248,
      "num_input_tokens_seen": 2097152,
      "step": 2
    },
    {
      "epoch": 0.0003145973154362416,
      "grad_norm": 98.5,
      "learning_rate": 5.042016806722689e-07,
      "loss": 7.9878,
      "num_input_tokens_seen": 3145728,
      "step": 3
    },
    {
      "epoch": 0.00041946308724832214,
      "grad_norm": 95.5,
      "learning_rate": 6.722689075630252e-07,
      "loss": 7.9434,
      "num_input_tokens_seen": 4194304,
      "step": 4
    }
  ],
  "logging_steps": 1,
  "max_steps": 9536,
  "num_input_tokens_seen": 4816896,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.094866300174336e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}