last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.014986886474334956,
  "eval_steps": 5,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0029973772948669914,
      "grad_norm": 1.7252535820007324,
      "learning_rate": 1e-05,
      "loss": 1.1987,
      "step": 1
    },
    {
      "epoch": 0.0029973772948669914,
      "eval_loss": 1.6100515127182007,
      "eval_runtime": 99.8256,
      "eval_samples_per_second": 5.63,
      "eval_steps_per_second": 2.815,
      "step": 1
    },
    {
      "epoch": 0.005994754589733983,
      "grad_norm": 1.6749064922332764,
      "learning_rate": 2e-05,
      "loss": 1.1175,
      "step": 2
    },
    {
      "epoch": 0.008992131884600974,
      "grad_norm": 1.857389211654663,
      "learning_rate": 3e-05,
      "loss": 1.2169,
      "step": 3
    },
    {
      "epoch": 0.011989509179467965,
      "grad_norm": 2.2108614444732666,
      "learning_rate": 4e-05,
      "loss": 1.4463,
      "step": 4
    },
    {
      "epoch": 0.014986886474334956,
      "grad_norm": 2.365875482559204,
      "learning_rate": 5e-05,
      "loss": 1.0604,
      "step": 5
    },
    {
      "epoch": 0.014986886474334956,
      "eval_loss": 1.3768054246902466,
      "eval_runtime": 13.669,
      "eval_samples_per_second": 41.115,
      "eval_steps_per_second": 20.558,
      "step": 5
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3745367602495488.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}