last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.14035087719298245,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.014035087719298246,
      "grad_norm": 0.07695987820625305,
      "learning_rate": 2e-05,
      "loss": 10.3829,
      "step": 1
    },
    {
      "epoch": 0.014035087719298246,
      "eval_loss": 10.382647514343262,
      "eval_runtime": 0.3354,
      "eval_samples_per_second": 89.442,
      "eval_steps_per_second": 44.721,
      "step": 1
    },
    {
      "epoch": 0.028070175438596492,
      "grad_norm": 0.06474371254444122,
      "learning_rate": 4e-05,
      "loss": 10.3801,
      "step": 2
    },
    {
      "epoch": 0.042105263157894736,
      "grad_norm": 0.06938017159700394,
      "learning_rate": 6e-05,
      "loss": 10.3815,
      "step": 3
    },
    {
      "epoch": 0.042105263157894736,
      "eval_loss": 10.382519721984863,
      "eval_runtime": 0.3334,
      "eval_samples_per_second": 89.995,
      "eval_steps_per_second": 44.997,
      "step": 3
    },
    {
      "epoch": 0.056140350877192984,
      "grad_norm": 0.06770727783441544,
      "learning_rate": 8e-05,
      "loss": 10.3813,
      "step": 4
    },
    {
      "epoch": 0.07017543859649122,
      "grad_norm": 0.07827406376600266,
      "learning_rate": 0.0001,
      "loss": 10.3822,
      "step": 5
    },
    {
      "epoch": 0.08421052631578947,
      "grad_norm": 0.06587807834148407,
      "learning_rate": 0.00012,
      "loss": 10.3815,
      "step": 6
    },
    {
      "epoch": 0.08421052631578947,
      "eval_loss": 10.38200569152832,
      "eval_runtime": 0.3582,
      "eval_samples_per_second": 83.759,
      "eval_steps_per_second": 41.879,
      "step": 6
    },
    {
      "epoch": 0.09824561403508772,
      "grad_norm": 0.06776691973209381,
      "learning_rate": 0.00014,
      "loss": 10.3802,
      "step": 7
    },
    {
      "epoch": 0.11228070175438597,
      "grad_norm": 0.07798208296298981,
      "learning_rate": 0.00016,
      "loss": 10.3819,
      "step": 8
    },
    {
      "epoch": 0.12631578947368421,
      "grad_norm": 0.08666855841875076,
      "learning_rate": 0.00018,
      "loss": 10.3803,
      "step": 9
    },
    {
      "epoch": 0.12631578947368421,
      "eval_loss": 10.381112098693848,
      "eval_runtime": 0.3302,
      "eval_samples_per_second": 90.852,
      "eval_steps_per_second": 45.426,
      "step": 9
    },
    {
      "epoch": 0.14035087719298245,
      "grad_norm": 0.07965846359729767,
      "learning_rate": 0.0002,
      "loss": 10.3806,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2086058065920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}