trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.920502092050209,
  "eval_steps": 500,
  "global_step": 87,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.33472803347280333,
      "grad_norm": 5.262314174920856,
      "learning_rate": 5e-06,
      "loss": 1.1175,
      "step": 10
    },
    {
      "epoch": 0.6694560669456067,
      "grad_norm": 5.981365790026769,
      "learning_rate": 5e-06,
      "loss": 1.0229,
      "step": 20
    },
    {
      "epoch": 0.9707112970711297,
      "eval_loss": 0.9536872506141663,
      "eval_runtime": 21.0481,
      "eval_samples_per_second": 38.103,
      "eval_steps_per_second": 0.618,
      "step": 29
    },
    {
      "epoch": 1.00836820083682,
      "grad_norm": 1.4208273735733241,
      "learning_rate": 5e-06,
      "loss": 0.9952,
      "step": 30
    },
    {
      "epoch": 1.3430962343096233,
      "grad_norm": 4.9064888353875835,
      "learning_rate": 5e-06,
      "loss": 0.9209,
      "step": 40
    },
    {
      "epoch": 1.6778242677824267,
      "grad_norm": 1.1284317077683494,
      "learning_rate": 5e-06,
      "loss": 0.8991,
      "step": 50
    },
    {
      "epoch": 1.9790794979079498,
      "eval_loss": 0.9046074151992798,
      "eval_runtime": 20.9928,
      "eval_samples_per_second": 38.204,
      "eval_steps_per_second": 0.619,
      "step": 59
    },
    {
      "epoch": 2.01673640167364,
      "grad_norm": 1.4649955809934467,
      "learning_rate": 5e-06,
      "loss": 0.9071,
      "step": 60
    },
    {
      "epoch": 2.3514644351464433,
      "grad_norm": 0.9844173290886701,
      "learning_rate": 5e-06,
      "loss": 0.8369,
      "step": 70
    },
    {
      "epoch": 2.6861924686192467,
      "grad_norm": 0.8203667838780383,
      "learning_rate": 5e-06,
      "loss": 0.8286,
      "step": 80
    },
    {
      "epoch": 2.920502092050209,
      "eval_loss": 0.8841589689254761,
      "eval_runtime": 19.9067,
      "eval_samples_per_second": 40.288,
      "eval_steps_per_second": 0.653,
      "step": 87
    },
    {
      "epoch": 2.920502092050209,
      "step": 87,
      "total_flos": 145518860697600.0,
      "train_loss": 0.931431375700852,
      "train_runtime": 3202.1178,
      "train_samples_per_second": 14.276,
      "train_steps_per_second": 0.027
    }
  ],
  "logging_steps": 10,
  "max_steps": 87,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 145518860697600.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}