last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0065930443382231745,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006593044338223175,
      "grad_norm": 14.93942928314209,
      "learning_rate": 2e-05,
      "loss": 8.2863,
      "step": 1
    },
    {
      "epoch": 0.0006593044338223175,
      "eval_loss": 2.058074712753296,
      "eval_runtime": 22.0546,
      "eval_samples_per_second": 28.974,
      "eval_steps_per_second": 14.509,
      "step": 1
    },
    {
      "epoch": 0.001318608867644635,
      "grad_norm": 18.237083435058594,
      "learning_rate": 4e-05,
      "loss": 7.8998,
      "step": 2
    },
    {
      "epoch": 0.0019779133014669525,
      "grad_norm": 15.265254974365234,
      "learning_rate": 6e-05,
      "loss": 8.0837,
      "step": 3
    },
    {
      "epoch": 0.0019779133014669525,
      "eval_loss": 2.02882981300354,
      "eval_runtime": 22.0496,
      "eval_samples_per_second": 28.98,
      "eval_steps_per_second": 14.513,
      "step": 3
    },
    {
      "epoch": 0.00263721773528927,
      "grad_norm": 13.398270606994629,
      "learning_rate": 8e-05,
      "loss": 6.9912,
      "step": 4
    },
    {
      "epoch": 0.0032965221691115872,
      "grad_norm": 13.810578346252441,
      "learning_rate": 0.0001,
      "loss": 6.9054,
      "step": 5
    },
    {
      "epoch": 0.003955826602933905,
      "grad_norm": 17.92094612121582,
      "learning_rate": 0.00012,
      "loss": 7.542,
      "step": 6
    },
    {
      "epoch": 0.003955826602933905,
      "eval_loss": 1.8383824825286865,
      "eval_runtime": 22.1925,
      "eval_samples_per_second": 28.794,
      "eval_steps_per_second": 14.419,
      "step": 6
    },
    {
      "epoch": 0.004615131036756222,
      "grad_norm": 12.2399320602417,
      "learning_rate": 0.00014,
      "loss": 6.4812,
      "step": 7
    },
    {
      "epoch": 0.00527443547057854,
      "grad_norm": 14.318899154663086,
      "learning_rate": 0.00016,
      "loss": 6.5268,
      "step": 8
    },
    {
      "epoch": 0.005933739904400857,
      "grad_norm": 17.184829711914062,
      "learning_rate": 0.00018,
      "loss": 6.084,
      "step": 9
    },
    {
      "epoch": 0.005933739904400857,
      "eval_loss": 1.6159027814865112,
      "eval_runtime": 22.1682,
      "eval_samples_per_second": 28.825,
      "eval_steps_per_second": 14.435,
      "step": 9
    },
    {
      "epoch": 0.0065930443382231745,
      "grad_norm": 16.13020133972168,
      "learning_rate": 0.0002,
      "loss": 6.8381,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1884120019894272.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}