trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 90,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 1.2691271906919943,
      "learning_rate": 5e-06,
      "loss": 0.6899,
      "step": 10
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.6356828917340551,
      "learning_rate": 5e-06,
      "loss": 0.5462,
      "step": 20
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.48838289017085196,
      "learning_rate": 5e-06,
      "loss": 0.5104,
      "step": 30
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.5004450082778931,
      "eval_runtime": 11.5822,
      "eval_samples_per_second": 69.762,
      "eval_steps_per_second": 1.122,
      "step": 30
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.4661895137762285,
      "learning_rate": 5e-06,
      "loss": 0.4644,
      "step": 40
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.457654242891002,
      "learning_rate": 5e-06,
      "loss": 0.4523,
      "step": 50
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.4551084466191181,
      "learning_rate": 5e-06,
      "loss": 0.4439,
      "step": 60
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.468788743019104,
      "eval_runtime": 11.8458,
      "eval_samples_per_second": 68.21,
      "eval_steps_per_second": 1.097,
      "step": 60
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 0.49518384341758054,
      "learning_rate": 5e-06,
      "loss": 0.4021,
      "step": 70
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.5289309565260419,
      "learning_rate": 5e-06,
      "loss": 0.3943,
      "step": 80
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.551959333275021,
      "learning_rate": 5e-06,
      "loss": 0.392,
      "step": 90
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.4573538899421692,
      "eval_runtime": 11.6616,
      "eval_samples_per_second": 69.287,
      "eval_steps_per_second": 1.115,
      "step": 90
    },
    {
      "epoch": 3.0,
      "step": 90,
      "total_flos": 150543972433920.0,
      "train_loss": 0.477274039056566,
      "train_runtime": 2568.8349,
      "train_samples_per_second": 17.919,
      "train_steps_per_second": 0.035
    }
  ],
  "logging_steps": 10,
  "max_steps": 90,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 150543972433920.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}