last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.14109347442680775,
  "eval_steps": 5,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02821869488536155,
      "grad_norm": 2.359462261199951,
      "learning_rate": 1e-05,
      "loss": 6.2205,
      "step": 1
    },
    {
      "epoch": 0.02821869488536155,
      "eval_loss": 6.2788615226745605,
      "eval_runtime": 36.398,
      "eval_samples_per_second": 1.648,
      "eval_steps_per_second": 0.824,
      "step": 1
    },
    {
      "epoch": 0.0564373897707231,
      "grad_norm": 2.0334458351135254,
      "learning_rate": 2e-05,
      "loss": 6.4757,
      "step": 2
    },
    {
      "epoch": 0.08465608465608465,
      "grad_norm": 2.242079973220825,
      "learning_rate": 3e-05,
      "loss": 6.3527,
      "step": 3
    },
    {
      "epoch": 0.1128747795414462,
      "grad_norm": 2.256972312927246,
      "learning_rate": 4e-05,
      "loss": 6.5204,
      "step": 4
    },
    {
      "epoch": 0.14109347442680775,
      "grad_norm": 2.1701977252960205,
      "learning_rate": 5e-05,
      "loss": 6.0037,
      "step": 5
    },
    {
      "epoch": 0.14109347442680775,
      "eval_loss": 6.136302471160889,
      "eval_runtime": 0.4876,
      "eval_samples_per_second": 123.04,
      "eval_steps_per_second": 61.52,
      "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": 511583956500480.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}