last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.18691588785046728,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.018691588785046728,
      "grad_norm": 1.2734662294387817,
      "learning_rate": 2e-05,
      "loss": 2.2598,
      "step": 1
    },
    {
      "epoch": 0.018691588785046728,
      "eval_loss": 2.2257795333862305,
      "eval_runtime": 0.9183,
      "eval_samples_per_second": 99.093,
      "eval_steps_per_second": 13.067,
      "step": 1
    },
    {
      "epoch": 0.037383177570093455,
      "grad_norm": 1.3796645402908325,
      "learning_rate": 4e-05,
      "loss": 2.262,
      "step": 2
    },
    {
      "epoch": 0.056074766355140186,
      "grad_norm": 1.3051254749298096,
      "learning_rate": 6e-05,
      "loss": 2.2198,
      "step": 3
    },
    {
      "epoch": 0.056074766355140186,
      "eval_loss": 2.2063558101654053,
      "eval_runtime": 0.9821,
      "eval_samples_per_second": 92.663,
      "eval_steps_per_second": 12.219,
      "step": 3
    },
    {
      "epoch": 0.07476635514018691,
      "grad_norm": 1.3184479475021362,
      "learning_rate": 8e-05,
      "loss": 2.1899,
      "step": 4
    },
    {
      "epoch": 0.09345794392523364,
      "grad_norm": 1.292141318321228,
      "learning_rate": 0.0001,
      "loss": 2.1399,
      "step": 5
    },
    {
      "epoch": 0.11214953271028037,
      "grad_norm": 1.3748526573181152,
      "learning_rate": 0.00012,
      "loss": 2.1182,
      "step": 6
    },
    {
      "epoch": 0.11214953271028037,
      "eval_loss": 1.8800315856933594,
      "eval_runtime": 0.9805,
      "eval_samples_per_second": 92.81,
      "eval_steps_per_second": 12.239,
      "step": 6
    },
    {
      "epoch": 0.1308411214953271,
      "grad_norm": 1.3897017240524292,
      "learning_rate": 0.00014,
      "loss": 1.8756,
      "step": 7
    },
    {
      "epoch": 0.14953271028037382,
      "grad_norm": 1.425822138786316,
      "learning_rate": 0.00016,
      "loss": 1.641,
      "step": 8
    },
    {
      "epoch": 0.16822429906542055,
      "grad_norm": 1.3553361892700195,
      "learning_rate": 0.00018,
      "loss": 1.7179,
      "step": 9
    },
    {
      "epoch": 0.16822429906542055,
      "eval_loss": 1.5405330657958984,
      "eval_runtime": 0.9796,
      "eval_samples_per_second": 92.894,
      "eval_steps_per_second": 12.25,
      "step": 9
    },
    {
      "epoch": 0.18691588785046728,
      "grad_norm": 1.5155949592590332,
      "learning_rate": 0.0002,
      "loss": 1.5614,
      "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": 7398256992583680.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}