last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.2777777777777778,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.009259259259259259,
      "eval_loss": NaN,
      "eval_runtime": 6.8496,
      "eval_samples_per_second": 6.716,
      "eval_steps_per_second": 3.358,
      "step": 1
    },
    {
      "epoch": 0.027777777777777776,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.046296296296296294,
      "eval_loss": NaN,
      "eval_runtime": 6.3513,
      "eval_samples_per_second": 7.243,
      "eval_steps_per_second": 3.621,
      "step": 5
    },
    {
      "epoch": 0.05555555555555555,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.08333333333333333,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.09259259259259259,
      "eval_loss": NaN,
      "eval_runtime": 6.3995,
      "eval_samples_per_second": 7.188,
      "eval_steps_per_second": 3.594,
      "step": 10
    },
    {
      "epoch": 0.1111111111111111,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.1388888888888889,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.1388888888888889,
      "eval_loss": NaN,
      "eval_runtime": 6.4311,
      "eval_samples_per_second": 7.153,
      "eval_steps_per_second": 3.576,
      "step": 15
    },
    {
      "epoch": 0.16666666666666666,
      "grad_norm": NaN,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.18518518518518517,
      "eval_loss": NaN,
      "eval_runtime": 6.4061,
      "eval_samples_per_second": 7.181,
      "eval_steps_per_second": 3.59,
      "step": 20
    },
    {
      "epoch": 0.19444444444444445,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": NaN,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.23148148148148148,
      "eval_loss": NaN,
      "eval_runtime": 6.4209,
      "eval_samples_per_second": 7.164,
      "eval_steps_per_second": 3.582,
      "step": 25
    },
    {
      "epoch": 0.25,
      "grad_norm": NaN,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.2777777777777778,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.2777777777777778,
      "eval_loss": NaN,
      "eval_runtime": 6.4337,
      "eval_samples_per_second": 7.15,
      "eval_steps_per_second": 3.575,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 15,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.06763800395776e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}