last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00046185643192813515,
  "eval_steps": 8,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.6185643192813516e-05,
      "eval_loss": 5.59415864944458,
      "eval_runtime": 1675.3459,
      "eval_samples_per_second": 5.442,
      "eval_steps_per_second": 2.721,
      "step": 1
    },
    {
      "epoch": 0.00013855692957844053,
      "grad_norm": 9.129727363586426,
      "learning_rate": 6e-05,
      "loss": 21.9235,
      "step": 3
    },
    {
      "epoch": 0.00027711385915688106,
      "grad_norm": 12.987714767456055,
      "learning_rate": 0.00012,
      "loss": 22.9083,
      "step": 6
    },
    {
      "epoch": 0.00036948514554250813,
      "eval_loss": 4.322031497955322,
      "eval_runtime": 1680.7058,
      "eval_samples_per_second": 5.425,
      "eval_steps_per_second": 2.713,
      "step": 8
    },
    {
      "epoch": 0.0004156707887353216,
      "grad_norm": 13.042325973510742,
      "learning_rate": 0.00018,
      "loss": 18.2704,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6347829962342400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}