checkpoint-9/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.3846153846153846,
  "eval_steps": 2,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 0.33971747756004333,
      "learning_rate": 2e-05,
      "loss": 1.2117,
      "step": 1
    },
    {
      "epoch": 0.15384615384615385,
      "eval_loss": 1.1688700914382935,
      "eval_runtime": 3.1352,
      "eval_samples_per_second": 15.948,
      "eval_steps_per_second": 0.319,
      "step": 1
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 0.34605106711387634,
      "learning_rate": 4e-05,
      "loss": 1.2196,
      "step": 2
    },
    {
      "epoch": 0.3076923076923077,
      "eval_loss": 1.1703702211380005,
      "eval_runtime": 3.1378,
      "eval_samples_per_second": 15.935,
      "eval_steps_per_second": 0.319,
      "step": 2
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 0.34318986535072327,
      "learning_rate": 6e-05,
      "loss": 1.2073,
      "step": 3
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.36031681299209595,
      "learning_rate": 8e-05,
      "loss": 1.1953,
      "step": 4
    },
    {
      "epoch": 0.6153846153846154,
      "eval_loss": 1.1325290203094482,
      "eval_runtime": 3.1573,
      "eval_samples_per_second": 15.837,
      "eval_steps_per_second": 0.317,
      "step": 4
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.3829631805419922,
      "learning_rate": 0.0001,
      "loss": 1.1648,
      "step": 5
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 0.3719262182712555,
      "learning_rate": 0.00012,
      "loss": 1.1059,
      "step": 6
    },
    {
      "epoch": 0.9230769230769231,
      "eval_loss": 0.987203061580658,
      "eval_runtime": 3.1404,
      "eval_samples_per_second": 15.921,
      "eval_steps_per_second": 0.318,
      "step": 6
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 0.33795467019081116,
      "learning_rate": 0.00014,
      "loss": 1.0255,
      "step": 7
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.22375954687595367,
      "learning_rate": 0.00016,
      "loss": 0.9508,
      "step": 8
    },
    {
      "epoch": 1.2307692307692308,
      "eval_loss": 0.9048197865486145,
      "eval_runtime": 3.1478,
      "eval_samples_per_second": 15.884,
      "eval_steps_per_second": 0.318,
      "step": 8
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.293130487203598,
      "learning_rate": 0.00018,
      "loss": 0.9355,
      "step": 9
    }
  ],
  "logging_steps": 1,
  "max_steps": 18,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.445530306702213e+17,
  "train_batch_size": 12,
  "trial_name": null,
  "trial_params": null
}