last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.000242666909333576,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.42666909333576e-05,
      "grad_norm": 8.384174346923828,
      "learning_rate": 2e-05,
      "loss": 13.288,
      "step": 1
    },
    {
      "epoch": 2.42666909333576e-05,
      "eval_loss": 3.190768003463745,
      "eval_runtime": 299.7529,
      "eval_samples_per_second": 57.884,
      "eval_steps_per_second": 28.944,
      "step": 1
    },
    {
      "epoch": 4.85333818667152e-05,
      "grad_norm": 5.884653091430664,
      "learning_rate": 4e-05,
      "loss": 13.1977,
      "step": 2
    },
    {
      "epoch": 7.28000728000728e-05,
      "grad_norm": 7.9959917068481445,
      "learning_rate": 6e-05,
      "loss": 13.2528,
      "step": 3
    },
    {
      "epoch": 7.28000728000728e-05,
      "eval_loss": 3.1809303760528564,
      "eval_runtime": 298.7554,
      "eval_samples_per_second": 58.078,
      "eval_steps_per_second": 29.04,
      "step": 3
    },
    {
      "epoch": 9.70667637334304e-05,
      "grad_norm": 9.162692070007324,
      "learning_rate": 8e-05,
      "loss": 13.8443,
      "step": 4
    },
    {
      "epoch": 0.000121333454666788,
      "grad_norm": 9.629426002502441,
      "learning_rate": 0.0001,
      "loss": 13.5097,
      "step": 5
    },
    {
      "epoch": 0.0001456001456001456,
      "grad_norm": 6.024376392364502,
      "learning_rate": 0.00012,
      "loss": 11.649,
      "step": 6
    },
    {
      "epoch": 0.0001456001456001456,
      "eval_loss": 3.0976181030273438,
      "eval_runtime": 299.6717,
      "eval_samples_per_second": 57.9,
      "eval_steps_per_second": 28.952,
      "step": 6
    },
    {
      "epoch": 0.0001698668365335032,
      "grad_norm": 4.765990734100342,
      "learning_rate": 0.00014,
      "loss": 11.1913,
      "step": 7
    },
    {
      "epoch": 0.0001941335274668608,
      "grad_norm": 11.631158828735352,
      "learning_rate": 0.00016,
      "loss": 14.2611,
      "step": 8
    },
    {
      "epoch": 0.0002184002184002184,
      "grad_norm": 6.4442057609558105,
      "learning_rate": 0.00018,
      "loss": 11.4726,
      "step": 9
    },
    {
      "epoch": 0.0002184002184002184,
      "eval_loss": 2.878862142562866,
      "eval_runtime": 299.2739,
      "eval_samples_per_second": 57.977,
      "eval_steps_per_second": 28.99,
      "step": 9
    },
    {
      "epoch": 0.000242666909333576,
      "grad_norm": 10.316732406616211,
      "learning_rate": 0.0002,
      "loss": 12.4694,
      "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": 669273056870400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}