last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.022922636103151862,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002292263610315186,
      "grad_norm": 8.022665977478027,
      "learning_rate": 2e-05,
      "loss": 2.6658,
      "step": 1
    },
    {
      "epoch": 0.002292263610315186,
      "eval_loss": 0.6854150891304016,
      "eval_runtime": 10.9071,
      "eval_samples_per_second": 16.87,
      "eval_steps_per_second": 8.435,
      "step": 1
    },
    {
      "epoch": 0.004584527220630372,
      "grad_norm": 9.221129417419434,
      "learning_rate": 4e-05,
      "loss": 2.8533,
      "step": 2
    },
    {
      "epoch": 0.006876790830945559,
      "grad_norm": 7.050643444061279,
      "learning_rate": 6e-05,
      "loss": 2.8236,
      "step": 3
    },
    {
      "epoch": 0.006876790830945559,
      "eval_loss": 0.6695802807807922,
      "eval_runtime": 10.3549,
      "eval_samples_per_second": 17.769,
      "eval_steps_per_second": 8.885,
      "step": 3
    },
    {
      "epoch": 0.009169054441260744,
      "grad_norm": 8.345035552978516,
      "learning_rate": 8e-05,
      "loss": 2.4904,
      "step": 4
    },
    {
      "epoch": 0.011461318051575931,
      "grad_norm": 7.962756633758545,
      "learning_rate": 0.0001,
      "loss": 2.3341,
      "step": 5
    },
    {
      "epoch": 0.013753581661891117,
      "grad_norm": 6.0530219078063965,
      "learning_rate": 0.00012,
      "loss": 2.4597,
      "step": 6
    },
    {
      "epoch": 0.013753581661891117,
      "eval_loss": 0.5594660043716431,
      "eval_runtime": 10.3706,
      "eval_samples_per_second": 17.742,
      "eval_steps_per_second": 8.871,
      "step": 6
    },
    {
      "epoch": 0.016045845272206302,
      "grad_norm": 6.4106974601745605,
      "learning_rate": 0.00014,
      "loss": 2.1572,
      "step": 7
    },
    {
      "epoch": 0.01833810888252149,
      "grad_norm": 4.761960983276367,
      "learning_rate": 0.00016,
      "loss": 1.9368,
      "step": 8
    },
    {
      "epoch": 0.020630372492836675,
      "grad_norm": 5.439733982086182,
      "learning_rate": 0.00018,
      "loss": 2.0602,
      "step": 9
    },
    {
      "epoch": 0.020630372492836675,
      "eval_loss": 0.49143391847610474,
      "eval_runtime": 10.3623,
      "eval_samples_per_second": 17.757,
      "eval_steps_per_second": 8.878,
      "step": 9
    },
    {
      "epoch": 0.022922636103151862,
      "grad_norm": 5.383951187133789,
      "learning_rate": 0.0002,
      "loss": 2.1832,
      "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": 3505343598428160.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}