last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0038763446070355654,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00038763446070355656,
      "grad_norm": 16.065155029296875,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 12.277,
      "step": 1
    },
    {
      "epoch": 0.00038763446070355656,
      "eval_loss": 12.859110832214355,
      "eval_runtime": 181.5941,
      "eval_samples_per_second": 5.986,
      "eval_steps_per_second": 2.996,
      "step": 1
    },
    {
      "epoch": 0.0007752689214071131,
      "grad_norm": 16.30716323852539,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 12.6827,
      "step": 2
    },
    {
      "epoch": 0.0011629033821106696,
      "grad_norm": 15.090091705322266,
      "learning_rate": 8.999999999999999e-05,
      "loss": 12.8142,
      "step": 3
    },
    {
      "epoch": 0.0011629033821106696,
      "eval_loss": 12.189451217651367,
      "eval_runtime": 182.2019,
      "eval_samples_per_second": 5.966,
      "eval_steps_per_second": 2.986,
      "step": 3
    },
    {
      "epoch": 0.0015505378428142262,
      "grad_norm": 17.527694702148438,
      "learning_rate": 0.00011999999999999999,
      "loss": 10.9965,
      "step": 4
    },
    {
      "epoch": 0.0019381723035177827,
      "grad_norm": 26.354629516601562,
      "learning_rate": 0.00015,
      "loss": 10.721,
      "step": 5
    },
    {
      "epoch": 0.002325806764221339,
      "grad_norm": 19.539798736572266,
      "learning_rate": 0.00017999999999999998,
      "loss": 6.8754,
      "step": 6
    },
    {
      "epoch": 0.002325806764221339,
      "eval_loss": 4.309875965118408,
      "eval_runtime": 182.3056,
      "eval_samples_per_second": 5.963,
      "eval_steps_per_second": 2.984,
      "step": 6
    },
    {
      "epoch": 0.002713441224924896,
      "grad_norm": 25.194347381591797,
      "learning_rate": 0.00020999999999999998,
      "loss": 4.5397,
      "step": 7
    },
    {
      "epoch": 0.0031010756856284525,
      "grad_norm": 20.890501022338867,
      "learning_rate": 0.00023999999999999998,
      "loss": 2.785,
      "step": 8
    },
    {
      "epoch": 0.003488710146332009,
      "grad_norm": 17.46876335144043,
      "learning_rate": 0.00027,
      "loss": 1.6347,
      "step": 9
    },
    {
      "epoch": 0.003488710146332009,
      "eval_loss": 1.2919782400131226,
      "eval_runtime": 182.2823,
      "eval_samples_per_second": 5.963,
      "eval_steps_per_second": 2.984,
      "step": 9
    },
    {
      "epoch": 0.0038763446070355654,
      "grad_norm": 22.62554931640625,
      "learning_rate": 0.0003,
      "loss": 1.3425,
      "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": 1.483774567120896e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}