{
  "best_metric": 0.7873479723930359,
  "best_model_checkpoint": "data/Llama-31-8B_task-1_120-samples_config-2_full_auto/checkpoint-60",
  "epoch": 18.0,
  "eval_steps": 500,
  "global_step": 99,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 1.7153618335723877,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.1531,
      "step": 1
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 1.8098291158676147,
      "learning_rate": 8.000000000000001e-06,
      "loss": 2.2152,
      "step": 2
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 1.3599419593811035,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 2.1572,
      "step": 4
    },
    {
      "epoch": 0.9090909090909091,
      "eval_loss": 2.114949941635132,
      "eval_runtime": 9.8884,
      "eval_samples_per_second": 2.427,
      "eval_steps_per_second": 2.427,
      "step": 5
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 1.332085371017456,
      "learning_rate": 2.4e-05,
      "loss": 2.1656,
      "step": 6
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 1.3887351751327515,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 2.0318,
      "step": 8
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 1.1428625583648682,
      "learning_rate": 4e-05,
      "loss": 2.0048,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.8149627447128296,
      "eval_runtime": 9.8912,
      "eval_samples_per_second": 2.426,
      "eval_steps_per_second": 2.426,
      "step": 11
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 1.609810709953308,
      "learning_rate": 4.8e-05,
      "loss": 1.7732,
      "step": 12
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 0.8945388197898865,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 1.7406,
      "step": 14
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 0.7094519734382629,
      "learning_rate": 6.400000000000001e-05,
      "loss": 1.651,
      "step": 16
    },
    {
      "epoch": 2.909090909090909,
      "eval_loss": 1.5523990392684937,
      "eval_runtime": 9.8912,
      "eval_samples_per_second": 2.426,
      "eval_steps_per_second": 2.426,
      "step": 16
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 0.7829838395118713,
      "learning_rate": 7.2e-05,
      "loss": 1.4713,
      "step": 18
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 0.8711206912994385,
      "learning_rate": 8e-05,
      "loss": 1.3717,
      "step": 20
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.9431599974632263,
      "learning_rate": 8.800000000000001e-05,
      "loss": 1.2397,
      "step": 22
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.1140679121017456,
      "eval_runtime": 9.8877,
      "eval_samples_per_second": 2.427,
      "eval_steps_per_second": 2.427,
      "step": 22
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 0.7719704508781433,
      "learning_rate": 9.6e-05,
      "loss": 1.0862,
      "step": 24
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 0.4783780872821808,
      "learning_rate": 9.999512620046522e-05,
      "loss": 0.9602,
      "step": 26
    },
    {
      "epoch": 4.909090909090909,
      "eval_loss": 0.9207101464271545,
      "eval_runtime": 9.8813,
      "eval_samples_per_second": 2.429,
      "eval_steps_per_second": 2.429,
      "step": 27
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 0.49044209718704224,
      "learning_rate": 9.995614150494293e-05,
      "loss": 0.8799,
      "step": 28
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 0.36221832036972046,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.8583,
      "step": 30
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 0.362589567899704,
      "learning_rate": 9.976136999909156e-05,
      "loss": 0.8646,
      "step": 32
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.8728799819946289,
      "eval_runtime": 9.8811,
      "eval_samples_per_second": 2.429,
      "eval_steps_per_second": 2.429,
      "step": 33
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 0.330808162689209,
      "learning_rate": 9.96057350657239e-05,
      "loss": 0.8273,
      "step": 34
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 0.3283514678478241,
      "learning_rate": 9.941141907232765e-05,
      "loss": 0.8784,
      "step": 36
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 0.3364407420158386,
      "learning_rate": 9.917857354066931e-05,
      "loss": 0.7895,
      "step": 38
    },
    {
      "epoch": 6.909090909090909,
      "eval_loss": 0.8490120768547058,
      "eval_runtime": 9.8857,
      "eval_samples_per_second": 2.428,
      "eval_steps_per_second": 2.428,
      "step": 38
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 0.3511122167110443,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.7966,
      "step": 40
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 0.36443227529525757,
      "learning_rate": 9.859805002892732e-05,
      "loss": 0.7329,
      "step": 42
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.3740652799606323,
      "learning_rate": 9.825082472361557e-05,
      "loss": 0.7762,
      "step": 44
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.8273000717163086,
      "eval_runtime": 9.8802,
      "eval_samples_per_second": 2.429,
      "eval_steps_per_second": 2.429,
      "step": 44
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 0.4024616479873657,
      "learning_rate": 9.786597487660337e-05,
      "loss": 0.7175,
      "step": 46
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 0.46190139651298523,
      "learning_rate": 9.744380058222483e-05,
      "loss": 0.7412,
      "step": 48
    },
    {
      "epoch": 8.909090909090908,
      "eval_loss": 0.8119735717773438,
      "eval_runtime": 9.8865,
      "eval_samples_per_second": 2.428,
      "eval_steps_per_second": 2.428,
      "step": 49
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 0.5113707184791565,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.7301,
      "step": 50
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 0.5442484021186829,
      "learning_rate": 9.648882429441257e-05,
      "loss": 0.7024,
      "step": 52
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 0.6121878027915955,
      "learning_rate": 9.595676696276172e-05,
      "loss": 0.6669,
      "step": 54
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.7971377968788147,
      "eval_runtime": 9.881,
      "eval_samples_per_second": 2.429,
      "eval_steps_per_second": 2.429,
      "step": 55
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 0.5720194578170776,
      "learning_rate": 9.538887392664544e-05,
      "loss": 0.6619,
      "step": 56
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 0.5266574621200562,
      "learning_rate": 9.478558801197065e-05,
      "loss": 0.6518,
      "step": 58
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 0.4223311245441437,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.6184,
      "step": 60
    },
    {
      "epoch": 10.909090909090908,
      "eval_loss": 0.7873479723930359,
      "eval_runtime": 9.8851,
      "eval_samples_per_second": 2.428,
      "eval_steps_per_second": 2.428,
      "step": 60
    },
    {
      "epoch": 11.272727272727273,
      "grad_norm": 0.4466268718242645,
      "learning_rate": 9.347474647526095e-05,
      "loss": 0.5785,
      "step": 62
    },
    {
      "epoch": 11.636363636363637,
      "grad_norm": 0.4064311683177948,
      "learning_rate": 9.276821300802534e-05,
      "loss": 0.5801,
      "step": 64
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.5084066987037659,
      "learning_rate": 9.202833017478422e-05,
      "loss": 0.5857,
      "step": 66
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.7943246364593506,
      "eval_runtime": 9.8798,
      "eval_samples_per_second": 2.429,
      "eval_steps_per_second": 2.429,
      "step": 66
    },
    {
      "epoch": 12.363636363636363,
      "grad_norm": 0.45895373821258545,
      "learning_rate": 9.125567491391476e-05,
      "loss": 0.5171,
      "step": 68
    },
    {
      "epoch": 12.727272727272727,
      "grad_norm": 0.494808167219162,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.5374,
      "step": 70
    },
    {
      "epoch": 12.909090909090908,
      "eval_loss": 0.8101968169212341,
      "eval_runtime": 9.8818,
      "eval_samples_per_second": 2.429,
      "eval_steps_per_second": 2.429,
      "step": 71
    },
    {
      "epoch": 13.090909090909092,
      "grad_norm": 0.5543341636657715,
      "learning_rate": 8.961448216775954e-05,
      "loss": 0.4999,
      "step": 72
    },
    {
      "epoch": 13.454545454545455,
      "grad_norm": 0.4971785843372345,
      "learning_rate": 8.874722443520899e-05,
      "loss": 0.4517,
      "step": 74
    },
    {
      "epoch": 13.818181818181818,
      "grad_norm": 0.622295081615448,
      "learning_rate": 8.784975278258783e-05,
      "loss": 0.4629,
      "step": 76
    },
    {
      "epoch": 14.0,
      "eval_loss": 0.8345339894294739,
      "eval_runtime": 9.8765,
      "eval_samples_per_second": 2.43,
      "eval_steps_per_second": 2.43,
      "step": 77
    },
    {
      "epoch": 14.181818181818182,
      "grad_norm": 0.6249837279319763,
      "learning_rate": 8.692276703129421e-05,
      "loss": 0.4408,
      "step": 78
    },
    {
      "epoch": 14.545454545454545,
      "grad_norm": 0.6758387088775635,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.4042,
      "step": 80
    },
    {
      "epoch": 14.909090909090908,
      "grad_norm": 0.7159420251846313,
      "learning_rate": 8.498316702566828e-05,
      "loss": 0.396,
      "step": 82
    },
    {
      "epoch": 14.909090909090908,
      "eval_loss": 0.8901812434196472,
      "eval_runtime": 9.928,
      "eval_samples_per_second": 2.417,
      "eval_steps_per_second": 2.417,
      "step": 82
    },
    {
      "epoch": 15.272727272727273,
      "grad_norm": 0.6551195383071899,
      "learning_rate": 8.397206521307584e-05,
      "loss": 0.3263,
      "step": 84
    },
    {
      "epoch": 15.636363636363637,
      "grad_norm": 0.7204640507698059,
      "learning_rate": 8.293447300593402e-05,
      "loss": 0.318,
      "step": 86
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.8874779939651489,
      "learning_rate": 8.18711994874345e-05,
      "loss": 0.336,
      "step": 88
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.9195945858955383,
      "eval_runtime": 9.8824,
      "eval_samples_per_second": 2.429,
      "eval_steps_per_second": 2.429,
      "step": 88
    },
    {
      "epoch": 16.363636363636363,
      "grad_norm": 0.8239272832870483,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.2693,
      "step": 90
    },
    {
      "epoch": 16.727272727272727,
      "grad_norm": 1.2460416555404663,
      "learning_rate": 7.967094433018508e-05,
      "loss": 0.2438,
      "step": 92
    },
    {
      "epoch": 16.90909090909091,
      "eval_loss": 1.0491713285446167,
      "eval_runtime": 9.8893,
      "eval_samples_per_second": 2.427,
      "eval_steps_per_second": 2.427,
      "step": 93
    },
    {
      "epoch": 17.09090909090909,
      "grad_norm": 0.9474222660064697,
      "learning_rate": 7.85356783842216e-05,
      "loss": 0.2363,
      "step": 94
    },
    {
      "epoch": 17.454545454545453,
      "grad_norm": 0.9264905452728271,
      "learning_rate": 7.737816117462752e-05,
      "loss": 0.1874,
      "step": 96
    },
    {
      "epoch": 17.818181818181817,
      "grad_norm": 1.1909377574920654,
      "learning_rate": 7.619929529850397e-05,
      "loss": 0.1943,
      "step": 98
    },
    {
      "epoch": 18.0,
      "eval_loss": 1.1072710752487183,
      "eval_runtime": 9.8799,
      "eval_samples_per_second": 2.429,
      "eval_steps_per_second": 2.429,
      "step": 99
    },
    {
      "epoch": 18.0,
      "step": 99,
      "total_flos": 2.635991088313139e+16,
      "train_loss": 0.8564288653809615,
      "train_runtime": 1910.1082,
      "train_samples_per_second": 2.304,
      "train_steps_per_second": 0.131
    }
  ],
  "logging_steps": 2,
  "max_steps": 250,
  "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": 2.635991088313139e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}