last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004715868898844612,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0004715868898844612,
      "grad_norm": 1.3327773809432983,
      "learning_rate": 2e-05,
      "loss": 2.2718,
      "step": 1
    },
    {
      "epoch": 0.0004715868898844612,
      "eval_loss": 1.7843847274780273,
      "eval_runtime": 17.041,
      "eval_samples_per_second": 52.403,
      "eval_steps_per_second": 26.231,
      "step": 1
    },
    {
      "epoch": 0.0009431737797689225,
      "grad_norm": 1.477419137954712,
      "learning_rate": 4e-05,
      "loss": 1.8746,
      "step": 2
    },
    {
      "epoch": 0.0014147606696533837,
      "grad_norm": 1.2436550855636597,
      "learning_rate": 6e-05,
      "loss": 1.7272,
      "step": 3
    },
    {
      "epoch": 0.0014147606696533837,
      "eval_loss": 1.7788474559783936,
      "eval_runtime": 15.7606,
      "eval_samples_per_second": 56.66,
      "eval_steps_per_second": 28.362,
      "step": 3
    },
    {
      "epoch": 0.001886347559537845,
      "grad_norm": 1.1900620460510254,
      "learning_rate": 8e-05,
      "loss": 1.63,
      "step": 4
    },
    {
      "epoch": 0.002357934449422306,
      "grad_norm": 0.9931256175041199,
      "learning_rate": 0.0001,
      "loss": 1.2157,
      "step": 5
    },
    {
      "epoch": 0.0028295213393067674,
      "grad_norm": 1.1468905210494995,
      "learning_rate": 0.00012,
      "loss": 1.5953,
      "step": 6
    },
    {
      "epoch": 0.0028295213393067674,
      "eval_loss": 1.71695077419281,
      "eval_runtime": 16.058,
      "eval_samples_per_second": 55.611,
      "eval_steps_per_second": 27.837,
      "step": 6
    },
    {
      "epoch": 0.0033011082291912286,
      "grad_norm": 1.1601518392562866,
      "learning_rate": 0.00014,
      "loss": 1.6881,
      "step": 7
    },
    {
      "epoch": 0.00377269511907569,
      "grad_norm": 1.0804229974746704,
      "learning_rate": 0.00016,
      "loss": 1.7769,
      "step": 8
    },
    {
      "epoch": 0.004244282008960151,
      "grad_norm": 1.4835208654403687,
      "learning_rate": 0.00018,
      "loss": 1.9097,
      "step": 9
    },
    {
      "epoch": 0.004244282008960151,
      "eval_loss": 1.6258902549743652,
      "eval_runtime": 15.9442,
      "eval_samples_per_second": 56.008,
      "eval_steps_per_second": 28.035,
      "step": 9
    },
    {
      "epoch": 0.004715868898844612,
      "grad_norm": 1.0217928886413574,
      "learning_rate": 0.0002,
      "loss": 1.4921,
      "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": 100167934279680.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}