last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0253324889170361,
  "eval_steps": 3,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00506649778340722,
      "grad_norm": 4.028384208679199,
      "learning_rate": 2e-05,
      "loss": 3.5544,
      "step": 1
    },
    {
      "epoch": 0.00506649778340722,
      "eval_loss": 3.5219967365264893,
      "eval_runtime": 17.0746,
      "eval_samples_per_second": 9.781,
      "eval_steps_per_second": 4.92,
      "step": 1
    },
    {
      "epoch": 0.01013299556681444,
      "grad_norm": 4.173046588897705,
      "learning_rate": 4e-05,
      "loss": 3.9016,
      "step": 2
    },
    {
      "epoch": 0.015199493350221659,
      "grad_norm": 3.8153645992279053,
      "learning_rate": 6e-05,
      "loss": 3.3707,
      "step": 3
    },
    {
      "epoch": 0.015199493350221659,
      "eval_loss": 3.4256696701049805,
      "eval_runtime": 17.0755,
      "eval_samples_per_second": 9.78,
      "eval_steps_per_second": 4.919,
      "step": 3
    },
    {
      "epoch": 0.02026599113362888,
      "grad_norm": 3.7728078365325928,
      "learning_rate": 8e-05,
      "loss": 3.2427,
      "step": 4
    },
    {
      "epoch": 0.0253324889170361,
      "grad_norm": 4.0927228927612305,
      "learning_rate": 0.0001,
      "loss": 2.8313,
      "step": 5
    }
  ],
  "logging_steps": 1,
  "max_steps": 25,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6610941168844800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}