{
  "best_metric": 2.204540491104126,
  "best_model_checkpoint": "./output/training_results/C014_llama3-8b-base_pretrain_20240428_005832/checkpoint-130",
  "epoch": 4.0,
  "eval_steps": 5,
  "global_step": 264,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.015151515151515152,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.5789,
      "step": 1
    },
    {
      "epoch": 0.015151515151515152,
      "eval_loss": 2.6458332538604736,
      "eval_runtime": 5.9609,
      "eval_samples_per_second": 78.008,
      "eval_steps_per_second": 0.671,
      "step": 1
    },
    {
      "epoch": 0.07575757575757576,
      "grad_norm": 3.5510944022733923,
      "learning_rate": 2.25e-06,
      "loss": 2.5672,
      "step": 5
    },
    {
      "epoch": 0.07575757575757576,
      "eval_loss": 2.628009080886841,
      "eval_runtime": 5.9819,
      "eval_samples_per_second": 77.735,
      "eval_steps_per_second": 0.669,
      "step": 5
    },
    {
      "epoch": 0.15151515151515152,
      "grad_norm": 3.473712979202145,
      "learning_rate": 6e-06,
      "loss": 2.5751,
      "step": 10
    },
    {
      "epoch": 0.15151515151515152,
      "eval_loss": 2.5313849449157715,
      "eval_runtime": 5.9441,
      "eval_samples_per_second": 78.228,
      "eval_steps_per_second": 0.673,
      "step": 10
    },
    {
      "epoch": 0.22727272727272727,
      "grad_norm": 2.7296451314204835,
      "learning_rate": 9.75e-06,
      "loss": 2.418,
      "step": 15
    },
    {
      "epoch": 0.22727272727272727,
      "eval_loss": 2.4634220600128174,
      "eval_runtime": 6.0122,
      "eval_samples_per_second": 77.343,
      "eval_steps_per_second": 0.665,
      "step": 15
    },
    {
      "epoch": 0.30303030303030304,
      "grad_norm": 2.9793222204470453,
      "learning_rate": 1.3500000000000001e-05,
      "loss": 2.4701,
      "step": 20
    },
    {
      "epoch": 0.30303030303030304,
      "eval_loss": 2.4176573753356934,
      "eval_runtime": 5.9735,
      "eval_samples_per_second": 77.844,
      "eval_steps_per_second": 0.67,
      "step": 20
    },
    {
      "epoch": 0.3787878787878788,
      "grad_norm": 2.9152379669367696,
      "learning_rate": 1.3097898548149108e-05,
      "loss": 2.3904,
      "step": 25
    },
    {
      "epoch": 0.3787878787878788,
      "eval_loss": 2.3785245418548584,
      "eval_runtime": 5.9994,
      "eval_samples_per_second": 77.507,
      "eval_steps_per_second": 0.667,
      "step": 25
    },
    {
      "epoch": 0.45454545454545453,
      "grad_norm": 2.475241901636818,
      "learning_rate": 1.041060545673204e-05,
      "loss": 2.3539,
      "step": 30
    },
    {
      "epoch": 0.45454545454545453,
      "eval_loss": 2.337780475616455,
      "eval_runtime": 5.9932,
      "eval_samples_per_second": 77.587,
      "eval_steps_per_second": 0.667,
      "step": 30
    },
    {
      "epoch": 0.5303030303030303,
      "grad_norm": 2.5312379436441272,
      "learning_rate": 8.236247706221891e-06,
      "loss": 2.3101,
      "step": 35
    },
    {
      "epoch": 0.5303030303030303,
      "eval_loss": 2.3082308769226074,
      "eval_runtime": 5.9901,
      "eval_samples_per_second": 77.628,
      "eval_steps_per_second": 0.668,
      "step": 35
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 2.5279244134684804,
      "learning_rate": 6.4849612135310325e-06,
      "loss": 2.3254,
      "step": 40
    },
    {
      "epoch": 0.6060606060606061,
      "eval_loss": 2.2816028594970703,
      "eval_runtime": 5.9798,
      "eval_samples_per_second": 77.762,
      "eval_steps_per_second": 0.669,
      "step": 40
    },
    {
      "epoch": 0.6818181818181818,
      "grad_norm": 2.3848893916874836,
      "learning_rate": 5.081159821297093e-06,
      "loss": 2.2762,
      "step": 45
    },
    {
      "epoch": 0.6818181818181818,
      "eval_loss": 2.2614095211029053,
      "eval_runtime": 5.9833,
      "eval_samples_per_second": 77.716,
      "eval_steps_per_second": 0.669,
      "step": 45
    },
    {
      "epoch": 0.7575757575757576,
      "grad_norm": 2.55130938777181,
      "learning_rate": 3.961509285889694e-06,
      "loss": 2.2525,
      "step": 50
    },
    {
      "epoch": 0.7575757575757576,
      "eval_loss": 2.2457971572875977,
      "eval_runtime": 6.0002,
      "eval_samples_per_second": 77.497,
      "eval_steps_per_second": 0.667,
      "step": 50
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 2.2003836383976156,
      "learning_rate": 3.073152889221908e-06,
      "loss": 2.2777,
      "step": 55
    },
    {
      "epoch": 0.8333333333333334,
      "eval_loss": 2.232052803039551,
      "eval_runtime": 5.9752,
      "eval_samples_per_second": 77.822,
      "eval_steps_per_second": 0.669,
      "step": 55
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 1.938340466235554,
      "learning_rate": 2.372162069694911e-06,
      "loss": 2.2054,
      "step": 60
    },
    {
      "epoch": 0.9090909090909091,
      "eval_loss": 2.2206437587738037,
      "eval_runtime": 5.9984,
      "eval_samples_per_second": 77.52,
      "eval_steps_per_second": 0.667,
      "step": 60
    },
    {
      "epoch": 0.9848484848484849,
      "grad_norm": 2.0152410097081375,
      "learning_rate": 1.8221877676625323e-06,
      "loss": 2.237,
      "step": 65
    },
    {
      "epoch": 0.9848484848484849,
      "eval_loss": 2.2112882137298584,
      "eval_runtime": 6.0081,
      "eval_samples_per_second": 77.396,
      "eval_steps_per_second": 0.666,
      "step": 65
    },
    {
      "epoch": 1.0606060606060606,
      "grad_norm": 2.206665008674913,
      "learning_rate": 1.3932903283558643e-06,
      "loss": 1.986,
      "step": 70
    },
    {
      "epoch": 1.0606060606060606,
      "eval_loss": 2.2115273475646973,
      "eval_runtime": 6.0121,
      "eval_samples_per_second": 77.344,
      "eval_steps_per_second": 0.665,
      "step": 70
    },
    {
      "epoch": 1.1363636363636362,
      "grad_norm": 2.4135250740210816,
      "learning_rate": 1.0609278071546894e-06,
      "loss": 1.9373,
      "step": 75
    },
    {
      "epoch": 1.1363636363636362,
      "eval_loss": 2.221705913543701,
      "eval_runtime": 5.966,
      "eval_samples_per_second": 77.942,
      "eval_steps_per_second": 0.67,
      "step": 75
    },
    {
      "epoch": 1.2121212121212122,
      "grad_norm": 2.0909203153994995,
      "learning_rate": 8.050843851687484e-07,
      "loss": 1.9228,
      "step": 80
    },
    {
      "epoch": 1.2121212121212122,
      "eval_loss": 2.2131617069244385,
      "eval_runtime": 6.0048,
      "eval_samples_per_second": 77.439,
      "eval_steps_per_second": 0.666,
      "step": 80
    },
    {
      "epoch": 1.2878787878787878,
      "grad_norm": 2.0701966451221723,
      "learning_rate": 6.095223338761627e-07,
      "loss": 1.9084,
      "step": 85
    },
    {
      "epoch": 1.2878787878787878,
      "eval_loss": 2.2117583751678467,
      "eval_runtime": 6.0144,
      "eval_samples_per_second": 77.314,
      "eval_steps_per_second": 0.665,
      "step": 85
    },
    {
      "epoch": 1.3636363636363638,
      "grad_norm": 2.0411250764985907,
      "learning_rate": 4.611425724763914e-07,
      "loss": 1.9684,
      "step": 90
    },
    {
      "epoch": 1.3636363636363638,
      "eval_loss": 2.212195634841919,
      "eval_runtime": 6.005,
      "eval_samples_per_second": 77.436,
      "eval_steps_per_second": 0.666,
      "step": 90
    },
    {
      "epoch": 1.4393939393939394,
      "grad_norm": 2.095523936778245,
      "learning_rate": 3.494403469094348e-07,
      "loss": 1.9126,
      "step": 95
    },
    {
      "epoch": 1.4393939393939394,
      "eval_loss": 2.2093794345855713,
      "eval_runtime": 5.9753,
      "eval_samples_per_second": 77.82,
      "eval_steps_per_second": 0.669,
      "step": 95
    },
    {
      "epoch": 1.5151515151515151,
      "grad_norm": 1.9491500689498569,
      "learning_rate": 2.660439312704735e-07,
      "loss": 1.9101,
      "step": 100
    },
    {
      "epoch": 1.5151515151515151,
      "eval_loss": 2.2066152095794678,
      "eval_runtime": 5.988,
      "eval_samples_per_second": 77.656,
      "eval_steps_per_second": 0.668,
      "step": 100
    },
    {
      "epoch": 1.5909090909090908,
      "grad_norm": 1.9297056004219728,
      "learning_rate": 2.0432551654866868e-07,
      "loss": 1.8496,
      "step": 105
    },
    {
      "epoch": 1.5909090909090908,
      "eval_loss": 2.2057933807373047,
      "eval_runtime": 6.029,
      "eval_samples_per_second": 77.127,
      "eval_steps_per_second": 0.663,
      "step": 105
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 2.0270269069156095,
      "learning_rate": 1.590746140201269e-07,
      "loss": 1.9154,
      "step": 110
    },
    {
      "epoch": 1.6666666666666665,
      "eval_loss": 2.205655097961426,
      "eval_runtime": 5.9865,
      "eval_samples_per_second": 77.675,
      "eval_steps_per_second": 0.668,
      "step": 110
    },
    {
      "epoch": 1.7424242424242424,
      "grad_norm": 2.0216556200111957,
      "learning_rate": 1.2622536684767967e-07,
      "loss": 1.9233,
      "step": 115
    },
    {
      "epoch": 1.7424242424242424,
      "eval_loss": 2.2055680751800537,
      "eval_runtime": 5.9969,
      "eval_samples_per_second": 77.54,
      "eval_steps_per_second": 0.667,
      "step": 115
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 2.121087680196771,
      "learning_rate": 1.0263013894441628e-07,
      "loss": 1.9198,
      "step": 120
    },
    {
      "epoch": 1.8181818181818183,
      "eval_loss": 2.205195665359497,
      "eval_runtime": 5.9814,
      "eval_samples_per_second": 77.741,
      "eval_steps_per_second": 0.669,
      "step": 120
    },
    {
      "epoch": 1.893939393939394,
      "grad_norm": 1.9887616789335223,
      "learning_rate": 8.587264024428055e-08,
      "loss": 1.9229,
      "step": 125
    },
    {
      "epoch": 1.893939393939394,
      "eval_loss": 2.2048099040985107,
      "eval_runtime": 6.0178,
      "eval_samples_per_second": 77.271,
      "eval_steps_per_second": 0.665,
      "step": 125
    },
    {
      "epoch": 1.9696969696969697,
      "grad_norm": 1.9737440679073275,
      "learning_rate": 7.411465733236604e-08,
      "loss": 1.8913,
      "step": 130
    },
    {
      "epoch": 1.9696969696969697,
      "eval_loss": 2.204540491104126,
      "eval_runtime": 6.0056,
      "eval_samples_per_second": 77.428,
      "eval_steps_per_second": 0.666,
      "step": 130
    },
    {
      "epoch": 2.0454545454545454,
      "grad_norm": 2.037256369037435,
      "learning_rate": 6.59711929010128e-08,
      "loss": 1.8814,
      "step": 135
    },
    {
      "epoch": 2.0454545454545454,
      "eval_loss": 2.204589605331421,
      "eval_runtime": 6.0067,
      "eval_samples_per_second": 77.414,
      "eval_steps_per_second": 0.666,
      "step": 135
    },
    {
      "epoch": 2.121212121212121,
      "grad_norm": 2.0713675635613353,
      "learning_rate": 6.040948153695873e-08,
      "loss": 1.8813,
      "step": 140
    },
    {
      "epoch": 2.121212121212121,
      "eval_loss": 2.2050745487213135,
      "eval_runtime": 5.9801,
      "eval_samples_per_second": 77.758,
      "eval_steps_per_second": 0.669,
      "step": 140
    },
    {
      "epoch": 2.196969696969697,
      "grad_norm": 1.9602573628685647,
      "learning_rate": 5.666794757151726e-08,
      "loss": 1.8912,
      "step": 145
    },
    {
      "epoch": 2.196969696969697,
      "eval_loss": 2.2057695388793945,
      "eval_runtime": 6.0053,
      "eval_samples_per_second": 77.431,
      "eval_steps_per_second": 0.666,
      "step": 145
    },
    {
      "epoch": 2.2727272727272725,
      "grad_norm": 2.0631081811496133,
      "learning_rate": 5.4191707642277796e-08,
      "loss": 1.9184,
      "step": 150
    },
    {
      "epoch": 2.2727272727272725,
      "eval_loss": 2.2065114974975586,
      "eval_runtime": 5.9555,
      "eval_samples_per_second": 78.079,
      "eval_steps_per_second": 0.672,
      "step": 150
    },
    {
      "epoch": 2.3484848484848486,
      "grad_norm": 2.0253318844317705,
      "learning_rate": 5.258170056372994e-08,
      "loss": 1.8662,
      "step": 155
    },
    {
      "epoch": 2.3484848484848486,
      "eval_loss": 2.207070827484131,
      "eval_runtime": 6.0047,
      "eval_samples_per_second": 77.439,
      "eval_steps_per_second": 0.666,
      "step": 155
    },
    {
      "epoch": 2.4242424242424243,
      "grad_norm": 2.073377995618171,
      "learning_rate": 5.1554954268425945e-08,
      "loss": 1.8809,
      "step": 160
    },
    {
      "epoch": 2.4242424242424243,
      "eval_loss": 2.2073893547058105,
      "eval_runtime": 6.0042,
      "eval_samples_per_second": 77.446,
      "eval_steps_per_second": 0.666,
      "step": 160
    },
    {
      "epoch": 2.5,
      "grad_norm": 1.9895760567683491,
      "learning_rate": 5.091387798309037e-08,
      "loss": 1.8591,
      "step": 165
    },
    {
      "epoch": 2.5,
      "eval_loss": 2.2076644897460938,
      "eval_runtime": 5.9917,
      "eval_samples_per_second": 77.608,
      "eval_steps_per_second": 0.668,
      "step": 165
    },
    {
      "epoch": 2.5757575757575757,
      "grad_norm": 2.0833747351029372,
      "learning_rate": 5.0522801309078135e-08,
      "loss": 1.8731,
      "step": 170
    },
    {
      "epoch": 2.5757575757575757,
      "eval_loss": 2.2079408168792725,
      "eval_runtime": 6.0146,
      "eval_samples_per_second": 77.312,
      "eval_steps_per_second": 0.665,
      "step": 170
    },
    {
      "epoch": 2.6515151515151514,
      "grad_norm": 2.0037327741523825,
      "learning_rate": 5.0290274187738543e-08,
      "loss": 1.8948,
      "step": 175
    },
    {
      "epoch": 2.6515151515151514,
      "eval_loss": 2.208183765411377,
      "eval_runtime": 6.0305,
      "eval_samples_per_second": 77.108,
      "eval_steps_per_second": 0.663,
      "step": 175
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 2.0305444519091917,
      "learning_rate": 5.015589639287439e-08,
      "loss": 1.8876,
      "step": 180
    },
    {
      "epoch": 2.7272727272727275,
      "eval_loss": 2.208235740661621,
      "eval_runtime": 5.9955,
      "eval_samples_per_second": 77.558,
      "eval_steps_per_second": 0.667,
      "step": 180
    },
    {
      "epoch": 2.8030303030303028,
      "grad_norm": 2.0062076076889266,
      "learning_rate": 5.0080665589248236e-08,
      "loss": 1.8408,
      "step": 185
    },
    {
      "epoch": 2.8030303030303028,
      "eval_loss": 2.2083210945129395,
      "eval_runtime": 6.0004,
      "eval_samples_per_second": 77.495,
      "eval_steps_per_second": 0.667,
      "step": 185
    },
    {
      "epoch": 2.878787878787879,
      "grad_norm": 2.043895270641295,
      "learning_rate": 5.004002235298783e-08,
      "loss": 1.8931,
      "step": 190
    },
    {
      "epoch": 2.878787878787879,
      "eval_loss": 2.208212375640869,
      "eval_runtime": 6.0002,
      "eval_samples_per_second": 77.498,
      "eval_steps_per_second": 0.667,
      "step": 190
    },
    {
      "epoch": 2.9545454545454546,
      "grad_norm": 1.9689184449811679,
      "learning_rate": 5.001893193212864e-08,
      "loss": 1.8569,
      "step": 195
    },
    {
      "epoch": 2.9545454545454546,
      "eval_loss": 2.2080307006835938,
      "eval_runtime": 5.9791,
      "eval_samples_per_second": 77.772,
      "eval_steps_per_second": 0.669,
      "step": 195
    },
    {
      "epoch": 3.0303030303030303,
      "grad_norm": 2.019410946950212,
      "learning_rate": 5.000847883910016e-08,
      "loss": 1.8621,
      "step": 200
    },
    {
      "epoch": 3.0303030303030303,
      "eval_loss": 2.207866907119751,
      "eval_runtime": 6.0003,
      "eval_samples_per_second": 77.496,
      "eval_steps_per_second": 0.667,
      "step": 200
    },
    {
      "epoch": 3.106060606060606,
      "grad_norm": 2.0283442915048644,
      "learning_rate": 5.000356435775757e-08,
      "loss": 1.8863,
      "step": 205
    },
    {
      "epoch": 3.106060606060606,
      "eval_loss": 2.207792043685913,
      "eval_runtime": 5.9858,
      "eval_samples_per_second": 77.684,
      "eval_steps_per_second": 0.668,
      "step": 205
    },
    {
      "epoch": 3.1818181818181817,
      "grad_norm": 2.05494798524839,
      "learning_rate": 5.0001391301969795e-08,
      "loss": 1.9021,
      "step": 210
    },
    {
      "epoch": 3.1818181818181817,
      "eval_loss": 2.2078535556793213,
      "eval_runtime": 5.9911,
      "eval_samples_per_second": 77.615,
      "eval_steps_per_second": 0.668,
      "step": 210
    },
    {
      "epoch": 3.257575757575758,
      "grad_norm": 2.0316955077847023,
      "learning_rate": 5.000049730753554e-08,
      "loss": 1.8648,
      "step": 215
    },
    {
      "epoch": 3.257575757575758,
      "eval_loss": 2.2079594135284424,
      "eval_runtime": 5.9692,
      "eval_samples_per_second": 77.9,
      "eval_steps_per_second": 0.67,
      "step": 215
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 2.0376583248479174,
      "learning_rate": 5.0000159841391415e-08,
      "loss": 1.8443,
      "step": 220
    },
    {
      "epoch": 3.3333333333333335,
      "eval_loss": 2.2080650329589844,
      "eval_runtime": 6.0237,
      "eval_samples_per_second": 77.195,
      "eval_steps_per_second": 0.664,
      "step": 220
    },
    {
      "epoch": 3.409090909090909,
      "grad_norm": 2.0309390490282864,
      "learning_rate": 5.0000045079130105e-08,
      "loss": 1.8978,
      "step": 225
    },
    {
      "epoch": 3.409090909090909,
      "eval_loss": 2.20800518989563,
      "eval_runtime": 6.0116,
      "eval_samples_per_second": 77.35,
      "eval_steps_per_second": 0.665,
      "step": 225
    },
    {
      "epoch": 3.484848484848485,
      "grad_norm": 2.0870431548602495,
      "learning_rate": 5.000001078153535e-08,
      "loss": 1.8658,
      "step": 230
    },
    {
      "epoch": 3.484848484848485,
      "eval_loss": 2.208038806915283,
      "eval_runtime": 5.99,
      "eval_samples_per_second": 77.629,
      "eval_steps_per_second": 0.668,
      "step": 230
    },
    {
      "epoch": 3.5606060606060606,
      "grad_norm": 2.0234951173098024,
      "learning_rate": 5.0000002081285866e-08,
      "loss": 1.8706,
      "step": 235
    },
    {
      "epoch": 3.5606060606060606,
      "eval_loss": 2.207921266555786,
      "eval_runtime": 5.9823,
      "eval_samples_per_second": 77.73,
      "eval_steps_per_second": 0.669,
      "step": 235
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 2.039329423643573,
      "learning_rate": 5.0000000300649115e-08,
      "loss": 1.8855,
      "step": 240
    },
    {
      "epoch": 3.6363636363636362,
      "eval_loss": 2.2077724933624268,
      "eval_runtime": 6.0052,
      "eval_samples_per_second": 77.433,
      "eval_steps_per_second": 0.666,
      "step": 240
    },
    {
      "epoch": 3.712121212121212,
      "grad_norm": 2.048524032150131,
      "learning_rate": 5.00000000286923e-08,
      "loss": 1.8535,
      "step": 245
    },
    {
      "epoch": 3.712121212121212,
      "eval_loss": 2.2078235149383545,
      "eval_runtime": 6.0225,
      "eval_samples_per_second": 77.21,
      "eval_steps_per_second": 0.664,
      "step": 245
    },
    {
      "epoch": 3.787878787878788,
      "grad_norm": 2.117736519999513,
      "learning_rate": 5.0000000001441026e-08,
      "loss": 1.9062,
      "step": 250
    },
    {
      "epoch": 3.787878787878788,
      "eval_loss": 2.207879066467285,
      "eval_runtime": 5.9714,
      "eval_samples_per_second": 77.871,
      "eval_steps_per_second": 0.67,
      "step": 250
    },
    {
      "epoch": 3.8636363636363638,
      "grad_norm": 2.0315594950889775,
      "learning_rate": 5.000000000002337e-08,
      "loss": 1.8628,
      "step": 255
    },
    {
      "epoch": 3.8636363636363638,
      "eval_loss": 2.2078306674957275,
      "eval_runtime": 6.0145,
      "eval_samples_per_second": 77.314,
      "eval_steps_per_second": 0.665,
      "step": 255
    },
    {
      "epoch": 3.9393939393939394,
      "grad_norm": 2.0132729461699994,
      "learning_rate": 5.0000000000000024e-08,
      "loss": 1.8484,
      "step": 260
    },
    {
      "epoch": 3.9393939393939394,
      "eval_loss": 2.207735776901245,
      "eval_runtime": 6.0067,
      "eval_samples_per_second": 77.413,
      "eval_steps_per_second": 0.666,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 264,
      "total_flos": 27428734894080.0,
      "train_loss": 2.005241002097274,
      "train_runtime": 8788.5675,
      "train_samples_per_second": 1.904,
      "train_steps_per_second": 0.03
    }
  ],
  "logging_steps": 5,
  "max_steps": 264,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 5,
  "total_flos": 27428734894080.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}