last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03748828491096532,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0037488284910965324,
      "grad_norm": 1.3008816242218018,
      "learning_rate": 2e-05,
      "loss": 3.046,
      "step": 1
    },
    {
      "epoch": 0.0037488284910965324,
      "eval_loss": 3.087820291519165,
      "eval_runtime": 5.8713,
      "eval_samples_per_second": 19.246,
      "eval_steps_per_second": 9.708,
      "step": 1
    },
    {
      "epoch": 0.007497656982193065,
      "grad_norm": 0.9666778445243835,
      "learning_rate": 4e-05,
      "loss": 3.0044,
      "step": 2
    },
    {
      "epoch": 0.011246485473289597,
      "grad_norm": 1.4649605751037598,
      "learning_rate": 6e-05,
      "loss": 3.4086,
      "step": 3
    },
    {
      "epoch": 0.011246485473289597,
      "eval_loss": 3.070920467376709,
      "eval_runtime": 5.8704,
      "eval_samples_per_second": 19.249,
      "eval_steps_per_second": 9.71,
      "step": 3
    },
    {
      "epoch": 0.01499531396438613,
      "grad_norm": 0.7611258625984192,
      "learning_rate": 8e-05,
      "loss": 2.6054,
      "step": 4
    },
    {
      "epoch": 0.01874414245548266,
      "grad_norm": 0.9841836094856262,
      "learning_rate": 0.0001,
      "loss": 2.8333,
      "step": 5
    },
    {
      "epoch": 0.022492970946579195,
      "grad_norm": 1.12917959690094,
      "learning_rate": 0.00012,
      "loss": 2.9573,
      "step": 6
    },
    {
      "epoch": 0.022492970946579195,
      "eval_loss": 2.9104862213134766,
      "eval_runtime": 5.8847,
      "eval_samples_per_second": 19.202,
      "eval_steps_per_second": 9.686,
      "step": 6
    },
    {
      "epoch": 0.026241799437675725,
      "grad_norm": 1.1534370183944702,
      "learning_rate": 0.00014,
      "loss": 2.6967,
      "step": 7
    },
    {
      "epoch": 0.02999062792877226,
      "grad_norm": 1.312908411026001,
      "learning_rate": 0.00016,
      "loss": 2.8173,
      "step": 8
    },
    {
      "epoch": 0.033739456419868794,
      "grad_norm": 1.73618745803833,
      "learning_rate": 0.00018,
      "loss": 2.93,
      "step": 9
    },
    {
      "epoch": 0.033739456419868794,
      "eval_loss": 2.74416184425354,
      "eval_runtime": 5.8873,
      "eval_samples_per_second": 19.194,
      "eval_steps_per_second": 9.682,
      "step": 9
    },
    {
      "epoch": 0.03748828491096532,
      "grad_norm": 1.6277101039886475,
      "learning_rate": 0.0002,
      "loss": 2.8048,
      "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": 2103703457759232.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}