checkpoint-200/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.32,
  "eval_steps": 50,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "learning_rate": 2.3869346733668342e-05,
      "loss": 1.2058,
      "step": 50
    },
    {
      "epoch": 0.08,
      "eval_loss": 0.9411209225654602,
      "eval_runtime": 405.7571,
      "eval_samples_per_second": 12.323,
      "eval_steps_per_second": 1.54,
      "step": 50
    },
    {
      "epoch": 0.16,
      "learning_rate": 2.2613065326633167e-05,
      "loss": 0.9065,
      "step": 100
    },
    {
      "epoch": 0.16,
      "eval_loss": 0.8835591673851013,
      "eval_runtime": 405.6798,
      "eval_samples_per_second": 12.325,
      "eval_steps_per_second": 1.541,
      "step": 100
    },
    {
      "epoch": 0.24,
      "learning_rate": 2.135678391959799e-05,
      "loss": 0.8612,
      "step": 150
    },
    {
      "epoch": 0.24,
      "eval_loss": 0.8560027480125427,
      "eval_runtime": 405.7296,
      "eval_samples_per_second": 12.323,
      "eval_steps_per_second": 1.54,
      "step": 150
    },
    {
      "epoch": 0.32,
      "learning_rate": 2.0100502512562815e-05,
      "loss": 0.8536,
      "step": 200
    },
    {
      "epoch": 0.32,
      "eval_loss": 0.8348749876022339,
      "eval_runtime": 405.5958,
      "eval_samples_per_second": 12.328,
      "eval_steps_per_second": 1.541,
      "step": 200
    }
  ],
  "logging_steps": 50,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 50,
  "total_flos": 3.50548150714368e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}