checkpoint-55/trainer_state.json

{
  "best_metric": 1.9656033515930176,
  "best_model_checkpoint": "outputs/checkpoint-55",
  "epoch": 4.782608695652174,
  "eval_steps": 500,
  "global_step": 55,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9565217391304348,
      "eval_loss": 2.8752284049987793,
      "eval_runtime": 34.3729,
      "eval_samples_per_second": 32.467,
      "eval_steps_per_second": 4.073,
      "step": 11
    },
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 1.172656774520874,
      "learning_rate": 2e-05,
      "loss": 2.8701,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.6693036556243896,
      "eval_runtime": 34.4065,
      "eval_samples_per_second": 32.436,
      "eval_steps_per_second": 4.069,
      "step": 23
    },
    {
      "epoch": 2.9565217391304346,
      "eval_loss": 2.1844232082366943,
      "eval_runtime": 34.3967,
      "eval_samples_per_second": 32.445,
      "eval_steps_per_second": 4.07,
      "step": 34
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 0.4842750132083893,
      "learning_rate": 4e-05,
      "loss": 2.5174,
      "step": 40
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.9845216274261475,
      "eval_runtime": 34.3962,
      "eval_samples_per_second": 32.445,
      "eval_steps_per_second": 4.07,
      "step": 46
    },
    {
      "epoch": 4.782608695652174,
      "eval_loss": 1.9656033515930176,
      "eval_runtime": 34.3887,
      "eval_samples_per_second": 32.452,
      "eval_steps_per_second": 4.071,
      "step": 55
    }
  ],
  "logging_steps": 20,
  "max_steps": 55,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 7693510509748224.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}