trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9182879377431905,
  "eval_steps": 500,
  "global_step": 7500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19,
      "learning_rate": 5e-05,
      "loss": 0.6919,
      "step": 500
    },
    {
      "epoch": 0.39,
      "learning_rate": 4.6532593619972264e-05,
      "loss": 0.6696,
      "step": 1000
    },
    {
      "epoch": 0.58,
      "learning_rate": 4.306518723994452e-05,
      "loss": 0.657,
      "step": 1500
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.959778085991678e-05,
      "loss": 0.6521,
      "step": 2000
    },
    {
      "epoch": 0.97,
      "learning_rate": 3.613037447988905e-05,
      "loss": 0.6467,
      "step": 2500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6400363445281982,
      "eval_runtime": 33.0485,
      "eval_samples_per_second": 311.027,
      "eval_steps_per_second": 4.872,
      "step": 2570
    },
    {
      "epoch": 1.17,
      "learning_rate": 3.266296809986131e-05,
      "loss": 0.6455,
      "step": 3000
    },
    {
      "epoch": 1.36,
      "learning_rate": 2.9195561719833564e-05,
      "loss": 0.6361,
      "step": 3500
    },
    {
      "epoch": 1.56,
      "learning_rate": 2.5728155339805826e-05,
      "loss": 0.6448,
      "step": 4000
    },
    {
      "epoch": 1.75,
      "learning_rate": 2.2260748959778087e-05,
      "loss": 0.6413,
      "step": 4500
    },
    {
      "epoch": 1.95,
      "learning_rate": 1.879334257975035e-05,
      "loss": 0.641,
      "step": 5000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6325713992118835,
      "eval_runtime": 33.21,
      "eval_samples_per_second": 309.515,
      "eval_steps_per_second": 4.848,
      "step": 5140
    },
    {
      "epoch": 2.14,
      "learning_rate": 1.5325936199722607e-05,
      "loss": 0.64,
      "step": 5500
    },
    {
      "epoch": 2.33,
      "learning_rate": 1.1858529819694868e-05,
      "loss": 0.6391,
      "step": 6000
    },
    {
      "epoch": 2.53,
      "learning_rate": 8.39112343966713e-06,
      "loss": 0.6442,
      "step": 6500
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.92371705963939e-06,
      "loss": 0.6312,
      "step": 7000
    },
    {
      "epoch": 2.92,
      "learning_rate": 1.4563106796116506e-06,
      "loss": 0.6375,
      "step": 7500
    }
  ],
  "logging_steps": 500,
  "max_steps": 7710,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1186351718870016.0,
  "trial_name": null,
  "trial_params": null
}