last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0011704804822379588,
  "eval_steps": 8,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00011704804822379587,
      "eval_loss": 4.554361820220947,
      "eval_runtime": 300.759,
      "eval_samples_per_second": 11.963,
      "eval_steps_per_second": 5.982,
      "step": 1
    },
    {
      "epoch": 0.0003511441446713876,
      "grad_norm": 4.222924709320068,
      "learning_rate": 6e-05,
      "loss": 4.4889,
      "step": 3
    },
    {
      "epoch": 0.0007022882893427752,
      "grad_norm": 5.726322650909424,
      "learning_rate": 0.00012,
      "loss": 4.3174,
      "step": 6
    },
    {
      "epoch": 0.000936384385790367,
      "eval_loss": 2.8311750888824463,
      "eval_runtime": 301.662,
      "eval_samples_per_second": 11.927,
      "eval_steps_per_second": 5.964,
      "step": 8
    },
    {
      "epoch": 0.001053432434014163,
      "grad_norm": 6.246539115905762,
      "learning_rate": 0.00018,
      "loss": 3.5675,
      "step": 9
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3709436417802240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}