last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.007416334476684898,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007416334476684898,
      "grad_norm": 12.539473533630371,
      "learning_rate": 2e-05,
      "loss": 18.8184,
      "step": 1
    },
    {
      "epoch": 0.0007416334476684898,
      "eval_loss": 2.656465768814087,
      "eval_runtime": 71.3819,
      "eval_samples_per_second": 7.957,
      "eval_steps_per_second": 7.957,
      "step": 1
    },
    {
      "epoch": 0.0014832668953369797,
      "grad_norm": 14.706053733825684,
      "learning_rate": 4e-05,
      "loss": 24.6349,
      "step": 2
    },
    {
      "epoch": 0.0022249003430054697,
      "grad_norm": 11.934353828430176,
      "learning_rate": 6e-05,
      "loss": 19.4363,
      "step": 3
    },
    {
      "epoch": 0.0022249003430054697,
      "eval_loss": 2.645974636077881,
      "eval_runtime": 71.0304,
      "eval_samples_per_second": 7.997,
      "eval_steps_per_second": 7.997,
      "step": 3
    },
    {
      "epoch": 0.0029665337906739594,
      "grad_norm": 12.361547470092773,
      "learning_rate": 8e-05,
      "loss": 19.716,
      "step": 4
    },
    {
      "epoch": 0.003708167238342449,
      "grad_norm": 20.54907989501953,
      "learning_rate": 0.0001,
      "loss": 20.2049,
      "step": 5
    },
    {
      "epoch": 0.004449800686010939,
      "grad_norm": 16.191509246826172,
      "learning_rate": 0.00012,
      "loss": 19.7239,
      "step": 6
    },
    {
      "epoch": 0.004449800686010939,
      "eval_loss": 2.478991985321045,
      "eval_runtime": 71.1391,
      "eval_samples_per_second": 7.984,
      "eval_steps_per_second": 7.984,
      "step": 6
    },
    {
      "epoch": 0.005191434133679429,
      "grad_norm": 27.116924285888672,
      "learning_rate": 0.00014,
      "loss": 16.9851,
      "step": 7
    },
    {
      "epoch": 0.005933067581347919,
      "grad_norm": 19.143783569335938,
      "learning_rate": 0.00016,
      "loss": 19.1227,
      "step": 8
    },
    {
      "epoch": 0.006674701029016409,
      "grad_norm": 26.77751922607422,
      "learning_rate": 0.00018,
      "loss": 20.1204,
      "step": 9
    },
    {
      "epoch": 0.006674701029016409,
      "eval_loss": 1.68876051902771,
      "eval_runtime": 71.0849,
      "eval_samples_per_second": 7.99,
      "eval_steps_per_second": 7.99,
      "step": 9
    },
    {
      "epoch": 0.007416334476684898,
      "grad_norm": 27.757686614990234,
      "learning_rate": 0.0002,
      "loss": 11.5992,
      "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": 813146528808960.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}