trainer_state.json

{
  "best_metric": 0.6320356726646423,
  "best_model_checkpoint": "./results/checkpoint-200",
  "epoch": 8.181818181818182,
  "eval_steps": 50,
  "global_step": 450,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.91,
      "learning_rate": 0.0002,
      "loss": 1.9348,
      "step": 50
    },
    {
      "epoch": 0.91,
      "eval_loss": 1.2767767906188965,
      "eval_runtime": 6.5658,
      "eval_samples_per_second": 3.808,
      "eval_steps_per_second": 0.609,
      "step": 50
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 0.8572,
      "step": 100
    },
    {
      "epoch": 1.82,
      "eval_loss": 0.7143042087554932,
      "eval_runtime": 6.9652,
      "eval_samples_per_second": 3.589,
      "eval_steps_per_second": 0.574,
      "step": 100
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0002,
      "loss": 0.5587,
      "step": 150
    },
    {
      "epoch": 2.73,
      "eval_loss": 0.6453649401664734,
      "eval_runtime": 6.6927,
      "eval_samples_per_second": 3.735,
      "eval_steps_per_second": 0.598,
      "step": 150
    },
    {
      "epoch": 3.64,
      "learning_rate": 0.0002,
      "loss": 0.4965,
      "step": 200
    },
    {
      "epoch": 3.64,
      "eval_loss": 0.6320356726646423,
      "eval_runtime": 7.193,
      "eval_samples_per_second": 3.476,
      "eval_steps_per_second": 0.556,
      "step": 200
    },
    {
      "epoch": 4.55,
      "learning_rate": 0.0002,
      "loss": 0.4535,
      "step": 250
    },
    {
      "epoch": 4.55,
      "eval_loss": 0.6632673144340515,
      "eval_runtime": 6.7989,
      "eval_samples_per_second": 3.677,
      "eval_steps_per_second": 0.588,
      "step": 250
    },
    {
      "epoch": 5.45,
      "learning_rate": 0.0002,
      "loss": 0.405,
      "step": 300
    },
    {
      "epoch": 5.45,
      "eval_loss": 0.6746855974197388,
      "eval_runtime": 6.7569,
      "eval_samples_per_second": 3.7,
      "eval_steps_per_second": 0.592,
      "step": 300
    },
    {
      "epoch": 6.36,
      "learning_rate": 0.0002,
      "loss": 0.4152,
      "step": 350
    },
    {
      "epoch": 6.36,
      "eval_loss": 0.7429562211036682,
      "eval_runtime": 6.9904,
      "eval_samples_per_second": 3.576,
      "eval_steps_per_second": 0.572,
      "step": 350
    },
    {
      "epoch": 7.27,
      "learning_rate": 0.0002,
      "loss": 0.3915,
      "step": 400
    },
    {
      "epoch": 7.27,
      "eval_loss": 0.6850793957710266,
      "eval_runtime": 7.0689,
      "eval_samples_per_second": 3.537,
      "eval_steps_per_second": 0.566,
      "step": 400
    },
    {
      "epoch": 8.18,
      "learning_rate": 0.0002,
      "loss": 0.3716,
      "step": 450
    },
    {
      "epoch": 8.18,
      "eval_loss": 0.6914016604423523,
      "eval_runtime": 6.7862,
      "eval_samples_per_second": 3.684,
      "eval_steps_per_second": 0.589,
      "step": 450
    }
  ],
  "logging_steps": 50,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 50,
  "total_flos": 83840921026560.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}