last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.008754650908295032,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008754650908295032,
      "grad_norm": 0.3211778402328491,
      "learning_rate": 2e-05,
      "loss": 0.9175,
      "step": 1
    },
    {
      "epoch": 0.0008754650908295032,
      "eval_loss": 1.0046578645706177,
      "eval_runtime": 46.7884,
      "eval_samples_per_second": 10.28,
      "eval_steps_per_second": 5.151,
      "step": 1
    },
    {
      "epoch": 0.0017509301816590063,
      "grad_norm": 0.3047505021095276,
      "learning_rate": 4e-05,
      "loss": 1.0556,
      "step": 2
    },
    {
      "epoch": 0.0026263952724885093,
      "grad_norm": 0.358992338180542,
      "learning_rate": 6e-05,
      "loss": 1.5189,
      "step": 3
    },
    {
      "epoch": 0.0026263952724885093,
      "eval_loss": 1.00363290309906,
      "eval_runtime": 45.4575,
      "eval_samples_per_second": 10.581,
      "eval_steps_per_second": 5.302,
      "step": 3
    },
    {
      "epoch": 0.0035018603633180127,
      "grad_norm": 0.3287069797515869,
      "learning_rate": 8e-05,
      "loss": 1.3581,
      "step": 4
    },
    {
      "epoch": 0.004377325454147516,
      "grad_norm": 0.3288503885269165,
      "learning_rate": 0.0001,
      "loss": 1.1164,
      "step": 5
    },
    {
      "epoch": 0.005252790544977019,
      "grad_norm": 0.43857482075691223,
      "learning_rate": 0.00012,
      "loss": 1.4435,
      "step": 6
    },
    {
      "epoch": 0.005252790544977019,
      "eval_loss": 0.9882233738899231,
      "eval_runtime": 45.5803,
      "eval_samples_per_second": 10.553,
      "eval_steps_per_second": 5.287,
      "step": 6
    },
    {
      "epoch": 0.006128255635806522,
      "grad_norm": 0.4593210816383362,
      "learning_rate": 0.00014,
      "loss": 1.3637,
      "step": 7
    },
    {
      "epoch": 0.007003720726636025,
      "grad_norm": 0.3701046407222748,
      "learning_rate": 0.00016,
      "loss": 0.7937,
      "step": 8
    },
    {
      "epoch": 0.007879185817465528,
      "grad_norm": 0.4593477249145508,
      "learning_rate": 0.00018,
      "loss": 0.5974,
      "step": 9
    },
    {
      "epoch": 0.007879185817465528,
      "eval_loss": 0.8962156176567078,
      "eval_runtime": 45.6067,
      "eval_samples_per_second": 10.547,
      "eval_steps_per_second": 5.284,
      "step": 9
    },
    {
      "epoch": 0.008754650908295032,
      "grad_norm": 0.3254682719707489,
      "learning_rate": 0.0002,
      "loss": 0.7293,
      "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": 6332410861977600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}