{
  "best_metric": 1.1641393899917603,
  "best_model_checkpoint": "data/Llama-31-8B_task-3_60-samples_config-2_full/checkpoint-48",
  "epoch": 24.0,
  "eval_steps": 500,
  "global_step": 69,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 0.5745601058006287,
      "learning_rate": 1e-05,
      "loss": 1.7172,
      "step": 1
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 0.47909945249557495,
      "learning_rate": 2e-05,
      "loss": 1.6831,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "eval_loss": 1.6899796724319458,
      "eval_runtime": 11.1827,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 2
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 0.4458988606929779,
      "learning_rate": 4e-05,
      "loss": 1.6856,
      "step": 4
    },
    {
      "epoch": 1.7391304347826086,
      "eval_loss": 1.6448274850845337,
      "eval_runtime": 11.1908,
      "eval_samples_per_second": 1.072,
      "eval_steps_per_second": 1.072,
      "step": 5
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 0.4629712402820587,
      "learning_rate": 6e-05,
      "loss": 1.671,
      "step": 6
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 0.467144638299942,
      "learning_rate": 8e-05,
      "loss": 1.6065,
      "step": 8
    },
    {
      "epoch": 2.782608695652174,
      "eval_loss": 1.5643361806869507,
      "eval_runtime": 11.1918,
      "eval_samples_per_second": 1.072,
      "eval_steps_per_second": 1.072,
      "step": 8
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 0.3479427695274353,
      "learning_rate": 0.0001,
      "loss": 1.5184,
      "step": 10
    },
    {
      "epoch": 3.8260869565217392,
      "eval_loss": 1.4830130338668823,
      "eval_runtime": 11.1894,
      "eval_samples_per_second": 1.072,
      "eval_steps_per_second": 1.072,
      "step": 11
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 0.23519960045814514,
      "learning_rate": 9.987820251299122e-05,
      "loss": 1.4882,
      "step": 12
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.24616719782352448,
      "learning_rate": 9.951340343707852e-05,
      "loss": 1.4227,
      "step": 14
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 1.4134116172790527,
      "eval_runtime": 11.1856,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 14
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 0.22191396355628967,
      "learning_rate": 9.890738003669029e-05,
      "loss": 1.3635,
      "step": 16
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 1.347265362739563,
      "eval_runtime": 11.1872,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 17
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 0.2585182785987854,
      "learning_rate": 9.806308479691595e-05,
      "loss": 1.3158,
      "step": 18
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 0.26426589488983154,
      "learning_rate": 9.698463103929542e-05,
      "loss": 1.2487,
      "step": 20
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 1.2793773412704468,
      "eval_runtime": 11.1842,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 20
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 0.22825951874256134,
      "learning_rate": 9.567727288213005e-05,
      "loss": 1.2196,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.2357500791549683,
      "eval_runtime": 11.1872,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 23
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 0.18481330573558807,
      "learning_rate": 9.414737964294636e-05,
      "loss": 1.1686,
      "step": 24
    },
    {
      "epoch": 8.695652173913043,
      "eval_loss": 1.218384027481079,
      "eval_runtime": 11.1871,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 25
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 0.18732167780399323,
      "learning_rate": 9.24024048078213e-05,
      "loss": 1.1427,
      "step": 26
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 0.1554836928844452,
      "learning_rate": 9.045084971874738e-05,
      "loss": 1.1271,
      "step": 28
    },
    {
      "epoch": 9.73913043478261,
      "eval_loss": 1.204110026359558,
      "eval_runtime": 11.187,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 28
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.1382986456155777,
      "learning_rate": 8.83022221559489e-05,
      "loss": 1.1329,
      "step": 30
    },
    {
      "epoch": 10.782608695652174,
      "eval_loss": 1.1925348043441772,
      "eval_runtime": 11.1891,
      "eval_samples_per_second": 1.072,
      "eval_steps_per_second": 1.072,
      "step": 31
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 0.15797151625156403,
      "learning_rate": 8.596699001693255e-05,
      "loss": 1.0772,
      "step": 32
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.1335517019033432,
      "learning_rate": 8.345653031794292e-05,
      "loss": 1.1049,
      "step": 34
    },
    {
      "epoch": 11.826086956521738,
      "eval_loss": 1.1837517023086548,
      "eval_runtime": 11.1916,
      "eval_samples_per_second": 1.072,
      "eval_steps_per_second": 1.072,
      "step": 34
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 0.13403163850307465,
      "learning_rate": 8.07830737662829e-05,
      "loss": 1.067,
      "step": 36
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 1.1764452457427979,
      "eval_runtime": 11.1842,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 37
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 0.14357908070087433,
      "learning_rate": 7.795964517353735e-05,
      "loss": 1.0779,
      "step": 38
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 0.1504034399986267,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.0693,
      "step": 40
    },
    {
      "epoch": 13.91304347826087,
      "eval_loss": 1.170665979385376,
      "eval_runtime": 11.1831,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 40
    },
    {
      "epoch": 14.608695652173914,
      "grad_norm": 0.13536736369132996,
      "learning_rate": 7.191855733945387e-05,
      "loss": 1.039,
      "step": 42
    },
    {
      "epoch": 14.956521739130435,
      "eval_loss": 1.1671864986419678,
      "eval_runtime": 11.1881,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 43
    },
    {
      "epoch": 15.304347826086957,
      "grad_norm": 0.1279880404472351,
      "learning_rate": 6.873032967079561e-05,
      "loss": 1.0399,
      "step": 44
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.15045784413814545,
      "learning_rate": 6.545084971874738e-05,
      "loss": 1.0381,
      "step": 46
    },
    {
      "epoch": 16.0,
      "eval_loss": 1.1651216745376587,
      "eval_runtime": 11.1971,
      "eval_samples_per_second": 1.072,
      "eval_steps_per_second": 1.072,
      "step": 46
    },
    {
      "epoch": 16.695652173913043,
      "grad_norm": 0.14994898438453674,
      "learning_rate": 6.209609477998338e-05,
      "loss": 0.994,
      "step": 48
    },
    {
      "epoch": 16.695652173913043,
      "eval_loss": 1.1641393899917603,
      "eval_runtime": 11.1852,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 48
    },
    {
      "epoch": 17.391304347826086,
      "grad_norm": 0.17326103150844574,
      "learning_rate": 5.868240888334653e-05,
      "loss": 1.0091,
      "step": 50
    },
    {
      "epoch": 17.73913043478261,
      "eval_loss": 1.1647818088531494,
      "eval_runtime": 11.1861,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 51
    },
    {
      "epoch": 18.08695652173913,
      "grad_norm": 0.15706367790699005,
      "learning_rate": 5.522642316338268e-05,
      "loss": 1.0247,
      "step": 52
    },
    {
      "epoch": 18.782608695652176,
      "grad_norm": 0.16293032467365265,
      "learning_rate": 5.174497483512506e-05,
      "loss": 0.996,
      "step": 54
    },
    {
      "epoch": 18.782608695652176,
      "eval_loss": 1.166734218597412,
      "eval_runtime": 11.1882,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 54
    },
    {
      "epoch": 19.47826086956522,
      "grad_norm": 0.18445666134357452,
      "learning_rate": 4.825502516487497e-05,
      "loss": 0.969,
      "step": 56
    },
    {
      "epoch": 19.82608695652174,
      "eval_loss": 1.1694756746292114,
      "eval_runtime": 11.1866,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 57
    },
    {
      "epoch": 20.17391304347826,
      "grad_norm": 0.1890542209148407,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.9787,
      "step": 58
    },
    {
      "epoch": 20.869565217391305,
      "grad_norm": 0.1912829726934433,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.9577,
      "step": 60
    },
    {
      "epoch": 20.869565217391305,
      "eval_loss": 1.1710036993026733,
      "eval_runtime": 11.1899,
      "eval_samples_per_second": 1.072,
      "eval_steps_per_second": 1.072,
      "step": 60
    },
    {
      "epoch": 21.565217391304348,
      "grad_norm": 0.2165202498435974,
      "learning_rate": 3.790390522001662e-05,
      "loss": 0.9489,
      "step": 62
    },
    {
      "epoch": 21.91304347826087,
      "eval_loss": 1.171953797340393,
      "eval_runtime": 11.1865,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 63
    },
    {
      "epoch": 22.26086956521739,
      "grad_norm": 0.21156741678714752,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.9391,
      "step": 64
    },
    {
      "epoch": 22.956521739130434,
      "grad_norm": 0.23810920119285583,
      "learning_rate": 3.12696703292044e-05,
      "loss": 0.9253,
      "step": 66
    },
    {
      "epoch": 22.956521739130434,
      "eval_loss": 1.176476001739502,
      "eval_runtime": 11.1876,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 66
    },
    {
      "epoch": 23.652173913043477,
      "grad_norm": 0.22348417341709137,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 0.9133,
      "step": 68
    },
    {
      "epoch": 24.0,
      "eval_loss": 1.180800199508667,
      "eval_runtime": 11.185,
      "eval_samples_per_second": 1.073,
      "eval_steps_per_second": 1.073,
      "step": 69
    },
    {
      "epoch": 24.0,
      "step": 69,
      "total_flos": 8.747351023812608e+16,
      "train_loss": 1.1726603568464085,
      "train_runtime": 3235.2716,
      "train_samples_per_second": 0.711,
      "train_steps_per_second": 0.031
    }
  ],
  "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": 8.747351023812608e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}