trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 7482,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 4.665864742047581e-05,
      "loss": 1.7562,
      "step": 500
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.331729484095162e-05,
      "loss": 1.526,
      "step": 1000
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.997594226142743e-05,
      "loss": 1.4549,
      "step": 1500
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.663458968190324e-05,
      "loss": 1.434,
      "step": 2000
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.3293237102379046e-05,
      "loss": 1.3931,
      "step": 2500
    },
    {
      "epoch": 0.4,
      "learning_rate": 2.995188452285485e-05,
      "loss": 1.3762,
      "step": 3000
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.6610531943330663e-05,
      "loss": 1.3459,
      "step": 3500
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.326917936380647e-05,
      "loss": 1.314,
      "step": 4000
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.9927826784282277e-05,
      "loss": 1.3203,
      "step": 4500
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.658647420475809e-05,
      "loss": 1.3109,
      "step": 5000
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.3245121625233894e-05,
      "loss": 1.3052,
      "step": 5500
    },
    {
      "epoch": 0.8,
      "learning_rate": 9.903769045709705e-06,
      "loss": 1.2921,
      "step": 6000
    },
    {
      "epoch": 0.87,
      "learning_rate": 6.562416466185513e-06,
      "loss": 1.2959,
      "step": 6500
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.2210638866613206e-06,
      "loss": 1.2986,
      "step": 7000
    },
    {
      "epoch": 1.0,
      "step": 7482,
      "total_flos": 2.662634158119936e+16,
      "train_loss": 1.3815043066249624,
      "train_runtime": 3193.6495,
      "train_samples_per_second": 4.686,
      "train_steps_per_second": 2.343
    }
  ],
  "logging_steps": 500,
  "max_steps": 7482,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 2.662634158119936e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}