last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0033297261300258055,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00033297261300258056,
      "grad_norm": 12.290924072265625,
      "learning_rate": 2e-05,
      "loss": 11.7566,
      "step": 1
    },
    {
      "epoch": 0.00033297261300258056,
      "eval_loss": 3.272779703140259,
      "eval_runtime": 16.6294,
      "eval_samples_per_second": 76.07,
      "eval_steps_per_second": 38.065,
      "step": 1
    },
    {
      "epoch": 0.0006659452260051611,
      "grad_norm": 17.70142364501953,
      "learning_rate": 4e-05,
      "loss": 15.1844,
      "step": 2
    },
    {
      "epoch": 0.0009989178390077415,
      "grad_norm": 13.565637588500977,
      "learning_rate": 6e-05,
      "loss": 13.606,
      "step": 3
    },
    {
      "epoch": 0.0009989178390077415,
      "eval_loss": 3.2558836936950684,
      "eval_runtime": 15.7601,
      "eval_samples_per_second": 80.266,
      "eval_steps_per_second": 40.165,
      "step": 3
    },
    {
      "epoch": 0.0013318904520103223,
      "grad_norm": 11.702593803405762,
      "learning_rate": 8e-05,
      "loss": 13.3459,
      "step": 4
    },
    {
      "epoch": 0.0016648630650129028,
      "grad_norm": 19.241825103759766,
      "learning_rate": 0.0001,
      "loss": 17.1201,
      "step": 5
    },
    {
      "epoch": 0.001997835678015483,
      "grad_norm": 10.895944595336914,
      "learning_rate": 0.00012,
      "loss": 12.4541,
      "step": 6
    },
    {
      "epoch": 0.001997835678015483,
      "eval_loss": 3.152925491333008,
      "eval_runtime": 15.7382,
      "eval_samples_per_second": 80.378,
      "eval_steps_per_second": 40.221,
      "step": 6
    },
    {
      "epoch": 0.0023308082910180636,
      "grad_norm": 15.67551040649414,
      "learning_rate": 0.00014,
      "loss": 10.4989,
      "step": 7
    },
    {
      "epoch": 0.0026637809040206445,
      "grad_norm": 13.790449142456055,
      "learning_rate": 0.00016,
      "loss": 13.007,
      "step": 8
    },
    {
      "epoch": 0.002996753517023225,
      "grad_norm": 14.053585052490234,
      "learning_rate": 0.00018,
      "loss": 13.0905,
      "step": 9
    },
    {
      "epoch": 0.002996753517023225,
      "eval_loss": 2.9347050189971924,
      "eval_runtime": 15.6529,
      "eval_samples_per_second": 80.816,
      "eval_steps_per_second": 40.44,
      "step": 9
    },
    {
      "epoch": 0.0033297261300258055,
      "grad_norm": 12.625859260559082,
      "learning_rate": 0.0002,
      "loss": 10.1853,
      "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": 77314746482688.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}