checkpoint-154/trainer_state.json

{
  "best_global_step": 150,
  "best_metric": 3.495365619659424,
  "best_model_checkpoint": "./control-qwen/checkpoint-100",
  "epoch": 14.0,
  "eval_steps": 50,
  "global_step": 154,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 98.99098205566406,
      "learning_rate": 6.25e-06,
      "loss": 61.474,
      "step": 10
    },
    {
      "epoch": 1.8470588235294119,
      "grad_norm": 125.53961944580078,
      "learning_rate": 9.943387951161702e-06,
      "loss": 54.0908,
      "step": 20
    },
    {
      "epoch": 2.7529411764705882,
      "grad_norm": 92.41641998291016,
      "learning_rate": 9.669186144114627e-06,
      "loss": 53.1726,
      "step": 30
    },
    {
      "epoch": 3.6588235294117646,
      "grad_norm": 66.3211898803711,
      "learning_rate": 9.179627397093184e-06,
      "loss": 42.4826,
      "step": 40
    },
    {
      "epoch": 4.564705882352941,
      "grad_norm": 48.61442565917969,
      "learning_rate": 8.497291635258235e-06,
      "loss": 39.6875,
      "step": 50
    },
    {
      "epoch": 4.564705882352941,
      "eval_loss": 6.142504692077637,
      "eval_runtime": 2.6504,
      "eval_samples_per_second": 3.773,
      "eval_steps_per_second": 3.773,
      "step": 50
    },
    {
      "epoch": 5.470588235294118,
      "grad_norm": 102.19329071044922,
      "learning_rate": 7.653650240809667e-06,
      "loss": 41.5856,
      "step": 60
    },
    {
      "epoch": 6.376470588235295,
      "grad_norm": 41.576393127441406,
      "learning_rate": 6.687614497970567e-06,
      "loss": 35.7429,
      "step": 70
    },
    {
      "epoch": 7.2823529411764705,
      "grad_norm": 28.25081443786621,
      "learning_rate": 5.643740887262905e-06,
      "loss": 30.2439,
      "step": 80
    },
    {
      "epoch": 8.188235294117646,
      "grad_norm": 60.2672119140625,
      "learning_rate": 4.570176006312769e-06,
      "loss": 30.5518,
      "step": 90
    },
    {
      "epoch": 9.094117647058823,
      "grad_norm": 52.88279724121094,
      "learning_rate": 3.5164359036325483e-06,
      "loss": 27.5296,
      "step": 100
    },
    {
      "epoch": 9.094117647058823,
      "eval_loss": 4.078804016113281,
      "eval_runtime": 2.648,
      "eval_samples_per_second": 3.776,
      "eval_steps_per_second": 3.776,
      "step": 100
    },
    {
      "epoch": 10.0,
      "grad_norm": 35.21394348144531,
      "learning_rate": 2.531122249200114e-06,
      "loss": 24.5956,
      "step": 110
    },
    {
      "epoch": 10.941176470588236,
      "grad_norm": 34.3149299621582,
      "learning_rate": 1.6596806789323317e-06,
      "loss": 25.0859,
      "step": 120
    },
    {
      "epoch": 11.847058823529412,
      "grad_norm": 24.832077026367188,
      "learning_rate": 9.423047049631956e-07,
      "loss": 21.9715,
      "step": 130
    },
    {
      "epoch": 12.75294117647059,
      "grad_norm": 30.87697410583496,
      "learning_rate": 4.1208186970303097e-07,
      "loss": 23.6025,
      "step": 140
    },
    {
      "epoch": 13.658823529411764,
      "grad_norm": 33.58473205566406,
      "learning_rate": 9.346764864195335e-08,
      "loss": 20.0079,
      "step": 150
    },
    {
      "epoch": 13.658823529411764,
      "eval_loss": 3.495365619659424,
      "eval_runtime": 2.6486,
      "eval_samples_per_second": 3.776,
      "eval_steps_per_second": 3.776,
      "step": 150
    }
  ],
  "logging_steps": 10,
  "max_steps": 154,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 14,
  "save_steps": 100,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 2,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.947828547971482e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}