punctuation/decoder_model/checkpoint-107/trainer_state.json

{
  "best_global_step": 107,
  "best_metric": 0.0014119020197540522,
  "best_model_checkpoint": "./models/punctuation/decoder_model/checkpoint-107",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 107,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09411764705882353,
      "grad_norm": 0.5588893890380859,
      "learning_rate": 1.0465116279069768e-05,
      "loss": 0.1531,
      "step": 10
    },
    {
      "epoch": 0.18823529411764706,
      "grad_norm": 0.3314439654350281,
      "learning_rate": 2.2093023255813955e-05,
      "loss": 0.0887,
      "step": 20
    },
    {
      "epoch": 0.2823529411764706,
      "grad_norm": 0.15414650738239288,
      "learning_rate": 3.372093023255814e-05,
      "loss": 0.03,
      "step": 30
    },
    {
      "epoch": 0.3764705882352941,
      "grad_norm": 0.11645973473787308,
      "learning_rate": 4.5348837209302326e-05,
      "loss": 0.0106,
      "step": 40
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 0.10382972657680511,
      "learning_rate": 4.997004260501766e-05,
      "loss": 0.0065,
      "step": 50
    },
    {
      "epoch": 0.5647058823529412,
      "grad_norm": 0.09259472787380219,
      "learning_rate": 4.978722957173222e-05,
      "loss": 0.0047,
      "step": 60
    },
    {
      "epoch": 0.6588235294117647,
      "grad_norm": 0.029875224456191063,
      "learning_rate": 4.943946161525192e-05,
      "loss": 0.0027,
      "step": 70
    },
    {
      "epoch": 0.7529411764705882,
      "grad_norm": 0.017065543681383133,
      "learning_rate": 4.8929053073949456e-05,
      "loss": 0.0017,
      "step": 80
    },
    {
      "epoch": 0.8470588235294118,
      "grad_norm": 0.11786513030529022,
      "learning_rate": 4.825940063245572e-05,
      "loss": 0.0027,
      "step": 90
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 0.027419880032539368,
      "learning_rate": 4.743496071728396e-05,
      "loss": 0.0024,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.0014119020197540522,
      "eval_model_preparation_time": 0.0128,
      "eval_runtime": 3.7995,
      "eval_samples_per_second": 26.319,
      "eval_steps_per_second": 13.16,
      "step": 107
    }
  ],
  "logging_steps": 10,
  "max_steps": 428,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3724928927141888e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}