last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.017921146953405017,
  "eval_steps": 5,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0035842293906810036,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 1.6449,
      "step": 1
    },
    {
      "epoch": 0.0035842293906810036,
      "eval_loss": 1.371159315109253,
      "eval_runtime": 8.0828,
      "eval_samples_per_second": 58.148,
      "eval_steps_per_second": 7.299,
      "step": 1
    },
    {
      "epoch": 0.007168458781362007,
      "grad_norm": 1.2307049036026,
      "learning_rate": 2e-05,
      "loss": 1.2685,
      "step": 2
    },
    {
      "epoch": 0.010752688172043012,
      "grad_norm": 1.0430188179016113,
      "learning_rate": 4e-05,
      "loss": 1.1567,
      "step": 3
    },
    {
      "epoch": 0.014336917562724014,
      "grad_norm": 1.1231369972229004,
      "learning_rate": 6e-05,
      "loss": 1.29,
      "step": 4
    },
    {
      "epoch": 0.017921146953405017,
      "grad_norm": 1.1865134239196777,
      "learning_rate": 8e-05,
      "loss": 1.2327,
      "step": 5
    },
    {
      "epoch": 0.017921146953405017,
      "eval_loss": 1.304574728012085,
      "eval_runtime": 8.1943,
      "eval_samples_per_second": 57.357,
      "eval_steps_per_second": 7.2,
      "step": 5
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4167081960407040.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}