last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0179640718562875,
  "eval_steps": 11,
  "global_step": 42,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.023952095808383235,
      "eval_loss": 0.9409148097038269,
      "eval_runtime": 0.4546,
      "eval_samples_per_second": 39.598,
      "eval_steps_per_second": 19.799,
      "step": 1
    },
    {
      "epoch": 0.11976047904191617,
      "grad_norm": 3.483912229537964,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.7711,
      "step": 5
    },
    {
      "epoch": 0.23952095808383234,
      "grad_norm": 5.300155162811279,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.9331,
      "step": 10
    },
    {
      "epoch": 0.2634730538922156,
      "eval_loss": 0.6188122630119324,
      "eval_runtime": 0.4408,
      "eval_samples_per_second": 40.839,
      "eval_steps_per_second": 20.42,
      "step": 11
    },
    {
      "epoch": 0.3592814371257485,
      "grad_norm": 4.500863075256348,
      "learning_rate": 5e-05,
      "loss": 0.428,
      "step": 15
    },
    {
      "epoch": 0.47904191616766467,
      "grad_norm": 1.3102383613586426,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.1093,
      "step": 20
    },
    {
      "epoch": 0.5269461077844312,
      "eval_loss": 0.007237435318529606,
      "eval_runtime": 0.4494,
      "eval_samples_per_second": 40.056,
      "eval_steps_per_second": 20.028,
      "step": 22
    },
    {
      "epoch": 0.5988023952095808,
      "grad_norm": 0.12453339993953705,
      "learning_rate": 8.333333333333334e-05,
      "loss": 0.0105,
      "step": 25
    },
    {
      "epoch": 0.718562874251497,
      "grad_norm": 0.22496868669986725,
      "learning_rate": 0.0001,
      "loss": 0.002,
      "step": 30
    },
    {
      "epoch": 0.7904191616766467,
      "eval_loss": 0.0012522797333076596,
      "eval_runtime": 0.4404,
      "eval_samples_per_second": 40.87,
      "eval_steps_per_second": 20.435,
      "step": 33
    },
    {
      "epoch": 0.8383233532934131,
      "grad_norm": 0.017963595688343048,
      "learning_rate": 6.294095225512603e-05,
      "loss": 0.0017,
      "step": 35
    },
    {
      "epoch": 0.9580838323353293,
      "grad_norm": 0.02263135462999344,
      "learning_rate": 6.698729810778065e-06,
      "loss": 0.0009,
      "step": 40
    }
  ],
  "logging_steps": 5,
  "max_steps": 42,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 747920575954944.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}