trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9963369963369964,
  "eval_steps": 500,
  "global_step": 51,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19536019536019536,
      "grad_norm": 461.63812255859375,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 2.882,
      "step": 10
    },
    {
      "epoch": 0.3907203907203907,
      "grad_norm": 21.538795471191406,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 3.1369,
      "step": 20
    },
    {
      "epoch": 0.5860805860805861,
      "grad_norm": 9.498225212097168,
      "learning_rate": 3e-06,
      "loss": 2.8571,
      "step": 30
    },
    {
      "epoch": 0.7814407814407814,
      "grad_norm": 14.18878173828125,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.879,
      "step": 40
    },
    {
      "epoch": 0.9768009768009768,
      "grad_norm": 19.329872131347656,
      "learning_rate": 5e-06,
      "loss": 2.7288,
      "step": 50
    }
  ],
  "logging_steps": 10,
  "max_steps": 51,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 313450454089728.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}