last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.013110455588331694,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0013110455588331695,
      "grad_norm": 0.7832544445991516,
      "learning_rate": 2e-05,
      "loss": 3.6267,
      "step": 1
    },
    {
      "epoch": 0.0013110455588331695,
      "eval_loss": 3.405561923980713,
      "eval_runtime": 15.1687,
      "eval_samples_per_second": 21.228,
      "eval_steps_per_second": 10.614,
      "step": 1
    },
    {
      "epoch": 0.002622091117666339,
      "grad_norm": 0.6407368183135986,
      "learning_rate": 4e-05,
      "loss": 3.5084,
      "step": 2
    },
    {
      "epoch": 0.003933136676499509,
      "grad_norm": 0.764837920665741,
      "learning_rate": 6e-05,
      "loss": 3.5129,
      "step": 3
    },
    {
      "epoch": 0.003933136676499509,
      "eval_loss": 3.403233528137207,
      "eval_runtime": 13.8314,
      "eval_samples_per_second": 23.28,
      "eval_steps_per_second": 11.64,
      "step": 3
    },
    {
      "epoch": 0.005244182235332678,
      "grad_norm": 0.6242678165435791,
      "learning_rate": 8e-05,
      "loss": 2.7061,
      "step": 4
    },
    {
      "epoch": 0.006555227794165847,
      "grad_norm": 0.9453555345535278,
      "learning_rate": 0.0001,
      "loss": 3.5085,
      "step": 5
    },
    {
      "epoch": 0.007866273352999017,
      "grad_norm": 0.7568758726119995,
      "learning_rate": 0.00012,
      "loss": 3.4138,
      "step": 6
    },
    {
      "epoch": 0.007866273352999017,
      "eval_loss": 3.364776372909546,
      "eval_runtime": 13.863,
      "eval_samples_per_second": 23.227,
      "eval_steps_per_second": 11.614,
      "step": 6
    },
    {
      "epoch": 0.009177318911832186,
      "grad_norm": 0.6948252320289612,
      "learning_rate": 0.00014,
      "loss": 3.1905,
      "step": 7
    },
    {
      "epoch": 0.010488364470665356,
      "grad_norm": 0.7294238209724426,
      "learning_rate": 0.00016,
      "loss": 3.0618,
      "step": 8
    },
    {
      "epoch": 0.011799410029498525,
      "grad_norm": 0.7105525135993958,
      "learning_rate": 0.00018,
      "loss": 2.8772,
      "step": 9
    },
    {
      "epoch": 0.011799410029498525,
      "eval_loss": 3.2387096881866455,
      "eval_runtime": 13.9114,
      "eval_samples_per_second": 23.147,
      "eval_steps_per_second": 11.573,
      "step": 9
    },
    {
      "epoch": 0.013110455588331694,
      "grad_norm": 0.7130061388015747,
      "learning_rate": 0.0002,
      "loss": 3.3902,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2918915377201152.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}