trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.84,
  "eval_steps": 500,
  "global_step": 24,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.32,
      "grad_norm": 0.20171859860420227,
      "learning_rate": 9.166666666666667e-05,
      "loss": 6.7889,
      "step": 2
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.2258663922548294,
      "learning_rate": 8.333333333333334e-05,
      "loss": 6.8518,
      "step": 4
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.2266591638326645,
      "learning_rate": 7.500000000000001e-05,
      "loss": 6.8562,
      "step": 6
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.19837072491645813,
      "learning_rate": 6.666666666666667e-05,
      "loss": 6.8282,
      "step": 8
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.17619071900844574,
      "learning_rate": 5.833333333333334e-05,
      "loss": 6.8528,
      "step": 10
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.12639087438583374,
      "learning_rate": 5e-05,
      "loss": 6.8231,
      "step": 12
    },
    {
      "epoch": 2.24,
      "grad_norm": 0.13687172532081604,
      "learning_rate": 4.166666666666667e-05,
      "loss": 6.7815,
      "step": 14
    },
    {
      "epoch": 2.56,
      "grad_norm": 0.18898847699165344,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 6.7792,
      "step": 16
    },
    {
      "epoch": 2.88,
      "grad_norm": 0.1648869514465332,
      "learning_rate": 2.5e-05,
      "loss": 6.8046,
      "step": 18
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.16822992265224457,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 6.8128,
      "step": 20
    },
    {
      "epoch": 3.52,
      "grad_norm": 0.16933274269104004,
      "learning_rate": 8.333333333333334e-06,
      "loss": 6.8081,
      "step": 22
    },
    {
      "epoch": 3.84,
      "grad_norm": 0.1729530692100525,
      "learning_rate": 0.0,
      "loss": 6.8425,
      "step": 24
    },
    {
      "epoch": 3.84,
      "step": 24,
      "total_flos": 92571267829680.0,
      "train_loss": 6.819143931070964,
      "train_runtime": 117.2771,
      "train_samples_per_second": 3.411,
      "train_steps_per_second": 0.205
    }
  ],
  "logging_steps": 2,
  "max_steps": 24,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 92571267829680.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}