last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.002153576956726563,
  "eval_steps": 4,
  "global_step": 8,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00026919711959082036,
      "grad_norm": 7.305245876312256,
      "learning_rate": 2e-05,
      "loss": 6.0527,
      "step": 1
    },
    {
      "epoch": 0.00026919711959082036,
      "eval_loss": 6.004375457763672,
      "eval_runtime": 186.7841,
      "eval_samples_per_second": 8.379,
      "eval_steps_per_second": 4.192,
      "step": 1
    },
    {
      "epoch": 0.0005383942391816407,
      "grad_norm": 6.773212432861328,
      "learning_rate": 4e-05,
      "loss": 6.151,
      "step": 2
    },
    {
      "epoch": 0.0008075913587724611,
      "grad_norm": 7.554437160491943,
      "learning_rate": 6e-05,
      "loss": 6.398,
      "step": 3
    },
    {
      "epoch": 0.0010767884783632815,
      "grad_norm": 6.408179759979248,
      "learning_rate": 8e-05,
      "loss": 5.7049,
      "step": 4
    },
    {
      "epoch": 0.0010767884783632815,
      "eval_loss": 5.3220696449279785,
      "eval_runtime": 187.1601,
      "eval_samples_per_second": 8.362,
      "eval_steps_per_second": 4.184,
      "step": 4
    },
    {
      "epoch": 0.0013459855979541019,
      "grad_norm": 6.113898754119873,
      "learning_rate": 0.0001,
      "loss": 5.5139,
      "step": 5
    },
    {
      "epoch": 0.0016151827175449223,
      "grad_norm": 4.3670973777771,
      "learning_rate": 0.00012,
      "loss": 4.8011,
      "step": 6
    },
    {
      "epoch": 0.0018843798371357427,
      "grad_norm": 8.779766082763672,
      "learning_rate": 0.00014,
      "loss": 4.9951,
      "step": 7
    },
    {
      "epoch": 0.002153576956726563,
      "grad_norm": 8.467183113098145,
      "learning_rate": 0.00016,
      "loss": 4.8208,
      "step": 8
    },
    {
      "epoch": 0.002153576956726563,
      "eval_loss": 4.797426700592041,
      "eval_runtime": 187.0275,
      "eval_samples_per_second": 8.368,
      "eval_steps_per_second": 4.187,
      "step": 8
    }
  ],
  "logging_steps": 1,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 4,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6135154892341248.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}