last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00013798194506248857,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.3798194506248858e-05,
      "grad_norm": 11.564373970031738,
      "learning_rate": 2e-05,
      "loss": 10.563,
      "step": 1
    },
    {
      "epoch": 1.3798194506248858e-05,
      "eval_loss": 10.16752815246582,
      "eval_runtime": 4125.8399,
      "eval_samples_per_second": 7.396,
      "eval_steps_per_second": 3.698,
      "step": 1
    },
    {
      "epoch": 2.7596389012497716e-05,
      "grad_norm": 13.19704532623291,
      "learning_rate": 4e-05,
      "loss": 11.0015,
      "step": 2
    },
    {
      "epoch": 4.139458351874657e-05,
      "grad_norm": 10.903892517089844,
      "learning_rate": 6e-05,
      "loss": 10.4111,
      "step": 3
    },
    {
      "epoch": 4.139458351874657e-05,
      "eval_loss": 9.961125373840332,
      "eval_runtime": 4123.3416,
      "eval_samples_per_second": 7.401,
      "eval_steps_per_second": 3.7,
      "step": 3
    },
    {
      "epoch": 5.519277802499543e-05,
      "grad_norm": 11.607290267944336,
      "learning_rate": 8e-05,
      "loss": 9.734,
      "step": 4
    },
    {
      "epoch": 6.899097253124429e-05,
      "grad_norm": 13.406005859375,
      "learning_rate": 0.0001,
      "loss": 7.917,
      "step": 5
    },
    {
      "epoch": 8.278916703749314e-05,
      "grad_norm": 12.174226760864258,
      "learning_rate": 0.00012,
      "loss": 8.65,
      "step": 6
    },
    {
      "epoch": 8.278916703749314e-05,
      "eval_loss": 7.667308807373047,
      "eval_runtime": 4120.9144,
      "eval_samples_per_second": 7.405,
      "eval_steps_per_second": 3.703,
      "step": 6
    },
    {
      "epoch": 9.6587361543742e-05,
      "grad_norm": 14.514153480529785,
      "learning_rate": 0.00014,
      "loss": 7.5376,
      "step": 7
    },
    {
      "epoch": 0.00011038555604999086,
      "grad_norm": 11.81023120880127,
      "learning_rate": 0.00016,
      "loss": 6.6161,
      "step": 8
    },
    {
      "epoch": 0.0001241837505562397,
      "grad_norm": 12.662259101867676,
      "learning_rate": 0.00018,
      "loss": 5.8399,
      "step": 9
    },
    {
      "epoch": 0.0001241837505562397,
      "eval_loss": 5.220765113830566,
      "eval_runtime": 4119.8507,
      "eval_samples_per_second": 7.407,
      "eval_steps_per_second": 3.704,
      "step": 9
    },
    {
      "epoch": 0.00013798194506248857,
      "grad_norm": 10.579679489135742,
      "learning_rate": 0.0002,
      "loss": 4.6917,
      "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": 1911091416268800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}