last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.007493443237167478,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007493443237167478,
      "grad_norm": 1.9591301679611206,
      "learning_rate": 2e-05,
      "loss": 1.1304,
      "step": 1
    },
    {
      "epoch": 0.0007493443237167478,
      "eval_loss": 1.6104313135147095,
      "eval_runtime": 18.4473,
      "eval_samples_per_second": 30.465,
      "eval_steps_per_second": 15.233,
      "step": 1
    },
    {
      "epoch": 0.0014986886474334957,
      "grad_norm": 2.605159044265747,
      "learning_rate": 4e-05,
      "loss": 1.5799,
      "step": 2
    },
    {
      "epoch": 0.0022480329711502436,
      "grad_norm": 1.5223053693771362,
      "learning_rate": 6e-05,
      "loss": 1.0511,
      "step": 3
    },
    {
      "epoch": 0.0022480329711502436,
      "eval_loss": 1.585645318031311,
      "eval_runtime": 17.0837,
      "eval_samples_per_second": 32.897,
      "eval_steps_per_second": 16.448,
      "step": 3
    },
    {
      "epoch": 0.0029973772948669914,
      "grad_norm": 2.100321054458618,
      "learning_rate": 8e-05,
      "loss": 1.1017,
      "step": 4
    },
    {
      "epoch": 0.003746721618583739,
      "grad_norm": 2.4209630489349365,
      "learning_rate": 0.0001,
      "loss": 1.0456,
      "step": 5
    },
    {
      "epoch": 0.004496065942300487,
      "grad_norm": 2.812591075897217,
      "learning_rate": 0.00012,
      "loss": 1.1463,
      "step": 6
    },
    {
      "epoch": 0.004496065942300487,
      "eval_loss": 1.1415724754333496,
      "eval_runtime": 17.1185,
      "eval_samples_per_second": 32.83,
      "eval_steps_per_second": 16.415,
      "step": 6
    },
    {
      "epoch": 0.005245410266017235,
      "grad_norm": 2.0659966468811035,
      "learning_rate": 0.00014,
      "loss": 0.7369,
      "step": 7
    },
    {
      "epoch": 0.005994754589733983,
      "grad_norm": 2.8209569454193115,
      "learning_rate": 0.00016,
      "loss": 0.6041,
      "step": 8
    },
    {
      "epoch": 0.0067440989134507304,
      "grad_norm": 7.074996471405029,
      "learning_rate": 0.00018,
      "loss": 0.6503,
      "step": 9
    },
    {
      "epoch": 0.0067440989134507304,
      "eval_loss": 0.537443995475769,
      "eval_runtime": 17.145,
      "eval_samples_per_second": 32.779,
      "eval_steps_per_second": 16.39,
      "step": 9
    },
    {
      "epoch": 0.007493443237167478,
      "grad_norm": 2.3110404014587402,
      "learning_rate": 0.0002,
      "loss": 0.3925,
      "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": 1895803354349568.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}