{
  "best_metric": 0.4068468511104584,
  "best_model_checkpoint": "data/Llama-31-8B_task-3_60-samples_config-2/checkpoint-31",
  "epoch": 17.73913043478261,
  "eval_steps": 500,
  "global_step": 51,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 2.713552474975586,
      "learning_rate": 1e-05,
      "loss": 2.4937,
      "step": 1
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 2.5405585765838623,
      "learning_rate": 2e-05,
      "loss": 2.4485,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "eval_loss": 2.4417779445648193,
      "eval_runtime": 11.1875,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 2
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 2.4669222831726074,
      "learning_rate": 4e-05,
      "loss": 2.3693,
      "step": 4
    },
    {
      "epoch": 1.7391304347826086,
      "eval_loss": 2.052797317504883,
      "eval_runtime": 11.1883,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 5
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 2.316904306411743,
      "learning_rate": 6e-05,
      "loss": 2.1298,
      "step": 6
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 2.0516061782836914,
      "learning_rate": 8e-05,
      "loss": 1.7056,
      "step": 8
    },
    {
      "epoch": 2.782608695652174,
      "eval_loss": 1.2637028694152832,
      "eval_runtime": 11.1901,
      "eval_samples_per_second": 1.072,
      "eval_steps_per_second": 1.072,
      "step": 8
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 1.4536932706832886,
      "learning_rate": 0.0001,
      "loss": 1.1524,
      "step": 10
    },
    {
      "epoch": 3.8260869565217392,
      "eval_loss": 0.6908593773841858,
      "eval_runtime": 11.1882,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 11
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 1.0161463022232056,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.6552,
      "step": 12
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.5411123633384705,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.568,
      "step": 14
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 0.5482725501060486,
      "eval_runtime": 11.1889,
      "eval_samples_per_second": 1.072,
      "eval_steps_per_second": 1.072,
      "step": 14
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 0.7326751351356506,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.5339,
      "step": 16
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 0.49588271975517273,
      "eval_runtime": 11.1842,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 17
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 0.7761788964271545,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.3388,
      "step": 18
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 0.476854532957077,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.4507,
      "step": 20
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 0.45868340134620667,
      "eval_runtime": 11.1861,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 20
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 0.8900018334388733,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.3073,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.44251060485839844,
      "eval_runtime": 11.1875,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 23
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 0.32006821036338806,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.4037,
      "step": 24
    },
    {
      "epoch": 8.695652173913043,
      "eval_loss": 0.4284121096134186,
      "eval_runtime": 11.1835,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 25
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 0.32192444801330566,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.3805,
      "step": 26
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 0.27814561128616333,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.3191,
      "step": 28
    },
    {
      "epoch": 9.73913043478261,
      "eval_loss": 0.4162331521511078,
      "eval_runtime": 11.1852,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 28
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.249674454331398,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.3306,
      "step": 30
    },
    {
      "epoch": 10.782608695652174,
      "eval_loss": 0.4068468511104584,
      "eval_runtime": 11.1894,
      "eval_samples_per_second": 1.072,
      "eval_steps_per_second": 1.072,
      "step": 31
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 0.3436458706855774,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.2947,
      "step": 32
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.5368538498878479,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.2342,
      "step": 34
    },
    {
      "epoch": 11.826086956521738,
      "eval_loss": 0.40730443596839905,
      "eval_runtime": 11.1855,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 34
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 0.4167219400405884,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.2395,
      "step": 36
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 0.42461928725242615,
      "eval_runtime": 11.1881,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 37
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 0.35160380601882935,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.2643,
      "step": 38
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 0.3089514374732971,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.1509,
      "step": 40
    },
    {
      "epoch": 13.91304347826087,
      "eval_loss": 0.43725764751434326,
      "eval_runtime": 11.1875,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 40
    },
    {
      "epoch": 14.608695652173914,
      "grad_norm": 0.4106989800930023,
      "learning_rate": 7.191855733945387e-05,
      "loss": 0.2333,
      "step": 42
    },
    {
      "epoch": 14.956521739130435,
      "eval_loss": 0.4344345033168793,
      "eval_runtime": 11.1853,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 43
    },
    {
      "epoch": 15.304347826086957,
      "grad_norm": 0.40442585945129395,
      "learning_rate": 6.873032967079561e-05,
      "loss": 0.1353,
      "step": 44
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.36868301033973694,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.1348,
      "step": 46
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.48776188492774963,
      "eval_runtime": 11.1846,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 46
    },
    {
      "epoch": 16.695652173913043,
      "grad_norm": 0.83082515001297,
      "learning_rate": 6.209609477998338e-05,
      "loss": 0.1198,
      "step": 48
    },
    {
      "epoch": 16.695652173913043,
      "eval_loss": 0.5003995299339294,
      "eval_runtime": 11.1878,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 48
    },
    {
      "epoch": 17.391304347826086,
      "grad_norm": 0.476839154958725,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.0922,
      "step": 50
    },
    {
      "epoch": 17.73913043478261,
      "eval_loss": 0.5109605193138123,
      "eval_runtime": 11.1872,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 51
    },
    {
      "epoch": 17.73913043478261,
      "step": 51,
      "total_flos": 6.560513253179392e+16,
      "train_loss": 0.630175527696516,
      "train_runtime": 2424.2514,
      "train_samples_per_second": 0.949,
      "train_steps_per_second": 0.041
    }
  ],
  "logging_steps": 2,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 25,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 7,
        "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": 6.560513253179392e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}