last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0015832387787951554,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 5.277462595983851e-05,
      "eval_loss": NaN,
      "eval_runtime": 918.9254,
      "eval_samples_per_second": 8.683,
      "eval_steps_per_second": 4.342,
      "step": 1
    },
    {
      "epoch": 0.00015832387787951554,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0002638731297991925,
      "eval_loss": NaN,
      "eval_runtime": 919.4135,
      "eval_samples_per_second": 8.678,
      "eval_steps_per_second": 4.34,
      "step": 5
    },
    {
      "epoch": 0.0003166477557590311,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.0004749716336385466,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.000527746259598385,
      "eval_loss": NaN,
      "eval_runtime": 919.387,
      "eval_samples_per_second": 8.679,
      "eval_steps_per_second": 4.34,
      "step": 10
    },
    {
      "epoch": 0.0006332955115180621,
      "grad_norm": NaN,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.0007916193893975777,
      "grad_norm": NaN,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0007916193893975777,
      "eval_loss": NaN,
      "eval_runtime": 1033.1093,
      "eval_samples_per_second": 7.723,
      "eval_steps_per_second": 3.862,
      "step": 15
    },
    {
      "epoch": 0.0009499432672770932,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.00105549251919677,
      "eval_loss": NaN,
      "eval_runtime": 1117.8965,
      "eval_samples_per_second": 7.138,
      "eval_steps_per_second": 3.569,
      "step": 20
    },
    {
      "epoch": 0.0011082671451566088,
      "grad_norm": NaN,
      "learning_rate": 8.435655349597689e-05,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.0012665910230361243,
      "grad_norm": NaN,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.0013193656489959627,
      "eval_loss": NaN,
      "eval_runtime": 919.4682,
      "eval_samples_per_second": 8.678,
      "eval_steps_per_second": 4.339,
      "step": 25
    },
    {
      "epoch": 0.0014249149009156398,
      "grad_norm": NaN,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.0015832387787951554,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.0015832387787951554,
      "eval_loss": NaN,
      "eval_runtime": 919.4392,
      "eval_samples_per_second": 8.678,
      "eval_steps_per_second": 4.34,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.6116868284678144e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}