last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.04292790313703908,
  "eval_steps": 10,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001100715465052284,
      "eval_loss": 0.370236873626709,
      "eval_runtime": 12.3987,
      "eval_samples_per_second": 30.89,
      "eval_steps_per_second": 15.486,
      "step": 1
    },
    {
      "epoch": 0.00550357732526142,
      "grad_norm": 0.7533979415893555,
      "learning_rate": 5e-05,
      "loss": 0.2584,
      "step": 5
    },
    {
      "epoch": 0.01100715465052284,
      "grad_norm": 0.6815402507781982,
      "learning_rate": 0.0001,
      "loss": 0.32,
      "step": 10
    },
    {
      "epoch": 0.01100715465052284,
      "eval_loss": 0.2777208089828491,
      "eval_runtime": 12.2381,
      "eval_samples_per_second": 31.296,
      "eval_steps_per_second": 15.689,
      "step": 10
    },
    {
      "epoch": 0.01651073197578426,
      "grad_norm": 0.19784417748451233,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.1716,
      "step": 15
    },
    {
      "epoch": 0.02201430930104568,
      "grad_norm": 0.503454327583313,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0933,
      "step": 20
    },
    {
      "epoch": 0.02201430930104568,
      "eval_loss": 0.08446875214576721,
      "eval_runtime": 12.3137,
      "eval_samples_per_second": 31.104,
      "eval_steps_per_second": 15.592,
      "step": 20
    },
    {
      "epoch": 0.0275178866263071,
      "grad_norm": 0.3515614867210388,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.05,
      "step": 25
    },
    {
      "epoch": 0.03302146395156852,
      "grad_norm": 0.31124940514564514,
      "learning_rate": 5e-05,
      "loss": 0.026,
      "step": 30
    },
    {
      "epoch": 0.03302146395156852,
      "eval_loss": 0.02826586551964283,
      "eval_runtime": 12.3902,
      "eval_samples_per_second": 30.912,
      "eval_steps_per_second": 15.496,
      "step": 30
    },
    {
      "epoch": 0.03852504127682994,
      "grad_norm": 0.21600860357284546,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0187,
      "step": 35
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1138274303213568.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}