last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.026917900403768506,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0026917900403768506,
      "grad_norm": 1.471656084060669,
      "learning_rate": 2e-05,
      "loss": 0.7396,
      "step": 1
    },
    {
      "epoch": 0.0026917900403768506,
      "eval_loss": 0.7381964921951294,
      "eval_runtime": 6.6874,
      "eval_samples_per_second": 11.813,
      "eval_steps_per_second": 11.813,
      "step": 1
    },
    {
      "epoch": 0.005383580080753701,
      "grad_norm": 1.4059834480285645,
      "learning_rate": 4e-05,
      "loss": 0.7273,
      "step": 2
    },
    {
      "epoch": 0.008075370121130552,
      "grad_norm": 1.7281467914581299,
      "learning_rate": 6e-05,
      "loss": 0.7709,
      "step": 3
    },
    {
      "epoch": 0.008075370121130552,
      "eval_loss": 0.6940876245498657,
      "eval_runtime": 6.0028,
      "eval_samples_per_second": 13.161,
      "eval_steps_per_second": 13.161,
      "step": 3
    },
    {
      "epoch": 0.010767160161507403,
      "grad_norm": 1.4571020603179932,
      "learning_rate": 8e-05,
      "loss": 0.7401,
      "step": 4
    },
    {
      "epoch": 0.013458950201884253,
      "grad_norm": 1.4379603862762451,
      "learning_rate": 0.0001,
      "loss": 0.7416,
      "step": 5
    },
    {
      "epoch": 0.016150740242261104,
      "grad_norm": 1.3622642755508423,
      "learning_rate": 0.00012,
      "loss": 0.6918,
      "step": 6
    },
    {
      "epoch": 0.016150740242261104,
      "eval_loss": 0.5574197769165039,
      "eval_runtime": 6.066,
      "eval_samples_per_second": 13.023,
      "eval_steps_per_second": 13.023,
      "step": 6
    },
    {
      "epoch": 0.018842530282637954,
      "grad_norm": 1.148888111114502,
      "learning_rate": 0.00014,
      "loss": 0.4991,
      "step": 7
    },
    {
      "epoch": 0.021534320323014805,
      "grad_norm": 1.1360676288604736,
      "learning_rate": 0.00016,
      "loss": 0.446,
      "step": 8
    },
    {
      "epoch": 0.024226110363391656,
      "grad_norm": 1.3324047327041626,
      "learning_rate": 0.00018,
      "loss": 0.4577,
      "step": 9
    },
    {
      "epoch": 0.024226110363391656,
      "eval_loss": 0.40684258937835693,
      "eval_runtime": 6.0984,
      "eval_samples_per_second": 12.954,
      "eval_steps_per_second": 12.954,
      "step": 9
    },
    {
      "epoch": 0.026917900403768506,
      "grad_norm": 1.2041003704071045,
      "learning_rate": 0.0002,
      "loss": 0.4258,
      "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": 2358194416386048.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}