last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.006207806316442927,
  "eval_steps": 5,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0012415612632885853,
      "grad_norm": 0.38651901483535767,
      "learning_rate": 1e-05,
      "loss": 0.8737,
      "step": 1
    },
    {
      "epoch": 0.0012415612632885853,
      "eval_loss": NaN,
      "eval_runtime": 71.5539,
      "eval_samples_per_second": 75.831,
      "eval_steps_per_second": 9.489,
      "step": 1
    },
    {
      "epoch": 0.0024831225265771707,
      "grad_norm": 0.3346116542816162,
      "learning_rate": 2e-05,
      "loss": 0.8453,
      "step": 2
    },
    {
      "epoch": 0.003724683789865756,
      "grad_norm": 0.34337764978408813,
      "learning_rate": 3e-05,
      "loss": 0.7958,
      "step": 3
    },
    {
      "epoch": 0.004966245053154341,
      "grad_norm": 0.4059295952320099,
      "learning_rate": 4e-05,
      "loss": 0.8395,
      "step": 4
    },
    {
      "epoch": 0.006207806316442927,
      "grad_norm": 0.3587191104888916,
      "learning_rate": 5e-05,
      "loss": 0.7908,
      "step": 5
    },
    {
      "epoch": 0.006207806316442927,
      "eval_loss": NaN,
      "eval_runtime": 60.0514,
      "eval_samples_per_second": 90.356,
      "eval_steps_per_second": 11.307,
      "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": 1.646990035255296e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}