{
  "best_metric": 1.5942710638046265,
  "best_model_checkpoint": "./output/training_results/C013_llama3-8b-base_pretrain_20240428_005832/checkpoint-65",
  "epoch": 4.0,
  "eval_steps": 5,
  "global_step": 268,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.014925373134328358,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.7594,
      "step": 1
    },
    {
      "epoch": 0.014925373134328358,
      "eval_loss": 1.7162834405899048,
      "eval_runtime": 5.9233,
      "eval_samples_per_second": 80.36,
      "eval_steps_per_second": 0.675,
      "step": 1
    },
    {
      "epoch": 0.07462686567164178,
      "grad_norm": 3.233517003168196,
      "learning_rate": 2.25e-06,
      "loss": 1.7333,
      "step": 5
    },
    {
      "epoch": 0.07462686567164178,
      "eval_loss": 1.7008264064788818,
      "eval_runtime": 5.9309,
      "eval_samples_per_second": 80.257,
      "eval_steps_per_second": 0.674,
      "step": 5
    },
    {
      "epoch": 0.14925373134328357,
      "grad_norm": 3.423591914627149,
      "learning_rate": 6e-06,
      "loss": 1.6854,
      "step": 10
    },
    {
      "epoch": 0.14925373134328357,
      "eval_loss": 1.682459831237793,
      "eval_runtime": 5.9184,
      "eval_samples_per_second": 80.427,
      "eval_steps_per_second": 0.676,
      "step": 10
    },
    {
      "epoch": 0.22388059701492538,
      "grad_norm": 2.75139227215463,
      "learning_rate": 9.75e-06,
      "loss": 1.6897,
      "step": 15
    },
    {
      "epoch": 0.22388059701492538,
      "eval_loss": 1.670121192932129,
      "eval_runtime": 5.9473,
      "eval_samples_per_second": 80.036,
      "eval_steps_per_second": 0.673,
      "step": 15
    },
    {
      "epoch": 0.29850746268656714,
      "grad_norm": 2.55059243594346,
      "learning_rate": 1.3500000000000001e-05,
      "loss": 1.6656,
      "step": 20
    },
    {
      "epoch": 0.29850746268656714,
      "eval_loss": 1.6651127338409424,
      "eval_runtime": 5.942,
      "eval_samples_per_second": 80.108,
      "eval_steps_per_second": 0.673,
      "step": 20
    },
    {
      "epoch": 0.373134328358209,
      "grad_norm": 2.7488276843880652,
      "learning_rate": 1.3725993373990923e-05,
      "loss": 1.7254,
      "step": 25
    },
    {
      "epoch": 0.373134328358209,
      "eval_loss": 1.6679344177246094,
      "eval_runtime": 5.9726,
      "eval_samples_per_second": 79.697,
      "eval_steps_per_second": 0.67,
      "step": 25
    },
    {
      "epoch": 0.44776119402985076,
      "grad_norm": 3.497477848836299,
      "learning_rate": 1.096088986215196e-05,
      "loss": 1.7178,
      "step": 30
    },
    {
      "epoch": 0.44776119402985076,
      "eval_loss": 1.6541944742202759,
      "eval_runtime": 5.953,
      "eval_samples_per_second": 79.96,
      "eval_steps_per_second": 0.672,
      "step": 30
    },
    {
      "epoch": 0.5223880597014925,
      "grad_norm": 2.7958991196951666,
      "learning_rate": 8.713726547724396e-06,
      "loss": 1.6656,
      "step": 35
    },
    {
      "epoch": 0.5223880597014925,
      "eval_loss": 1.6459277868270874,
      "eval_runtime": 5.9488,
      "eval_samples_per_second": 80.016,
      "eval_steps_per_second": 0.672,
      "step": 35
    },
    {
      "epoch": 0.5970149253731343,
      "grad_norm": 2.2395806127723907,
      "learning_rate": 6.8954890860223175e-06,
      "loss": 1.6647,
      "step": 40
    },
    {
      "epoch": 0.5970149253731343,
      "eval_loss": 1.6308344602584839,
      "eval_runtime": 5.9441,
      "eval_samples_per_second": 80.08,
      "eval_steps_per_second": 0.673,
      "step": 40
    },
    {
      "epoch": 0.6716417910447762,
      "grad_norm": 2.542267170825853,
      "learning_rate": 5.431031497828169e-06,
      "loss": 1.6645,
      "step": 45
    },
    {
      "epoch": 0.6716417910447762,
      "eval_loss": 1.6204748153686523,
      "eval_runtime": 5.9555,
      "eval_samples_per_second": 79.927,
      "eval_steps_per_second": 0.672,
      "step": 45
    },
    {
      "epoch": 0.746268656716418,
      "grad_norm": 2.796577480411167,
      "learning_rate": 4.257147264863097e-06,
      "loss": 1.6151,
      "step": 50
    },
    {
      "epoch": 0.746268656716418,
      "eval_loss": 1.6129424571990967,
      "eval_runtime": 5.9792,
      "eval_samples_per_second": 79.609,
      "eval_steps_per_second": 0.669,
      "step": 50
    },
    {
      "epoch": 0.8208955223880597,
      "grad_norm": 2.8952316281119757,
      "learning_rate": 3.320875141743125e-06,
      "loss": 1.6359,
      "step": 55
    },
    {
      "epoch": 0.8208955223880597,
      "eval_loss": 1.6052401065826416,
      "eval_runtime": 5.9501,
      "eval_samples_per_second": 79.998,
      "eval_steps_per_second": 0.672,
      "step": 55
    },
    {
      "epoch": 0.8955223880597015,
      "grad_norm": 2.455450911279608,
      "learning_rate": 2.57801538033309e-06,
      "loss": 1.5885,
      "step": 60
    },
    {
      "epoch": 0.8955223880597015,
      "eval_loss": 1.5995320081710815,
      "eval_runtime": 5.9574,
      "eval_samples_per_second": 79.9,
      "eval_steps_per_second": 0.671,
      "step": 60
    },
    {
      "epoch": 0.9701492537313433,
      "grad_norm": 1.8236330059926131,
      "learning_rate": 1.9918341191886553e-06,
      "loss": 1.6142,
      "step": 65
    },
    {
      "epoch": 0.9701492537313433,
      "eval_loss": 1.5942710638046265,
      "eval_runtime": 5.9491,
      "eval_samples_per_second": 80.013,
      "eval_steps_per_second": 0.672,
      "step": 65
    },
    {
      "epoch": 1.044776119402985,
      "grad_norm": 2.5122250154998893,
      "learning_rate": 1.5319356141270557e-06,
      "loss": 1.4875,
      "step": 70
    },
    {
      "epoch": 1.044776119402985,
      "eval_loss": 1.5963352918624878,
      "eval_runtime": 5.9465,
      "eval_samples_per_second": 80.047,
      "eval_steps_per_second": 0.673,
      "step": 70
    },
    {
      "epoch": 1.1194029850746268,
      "grad_norm": 2.2374216929111017,
      "learning_rate": 1.173283781790582e-06,
      "loss": 1.3844,
      "step": 75
    },
    {
      "epoch": 1.1194029850746268,
      "eval_loss": 1.611823558807373,
      "eval_runtime": 5.9845,
      "eval_samples_per_second": 79.538,
      "eval_steps_per_second": 0.668,
      "step": 75
    },
    {
      "epoch": 1.1940298507462686,
      "grad_norm": 2.061088772790073,
      "learning_rate": 8.95356202130823e-07,
      "loss": 1.3555,
      "step": 80
    },
    {
      "epoch": 1.1940298507462686,
      "eval_loss": 1.6069068908691406,
      "eval_runtime": 5.9508,
      "eval_samples_per_second": 79.99,
      "eval_steps_per_second": 0.672,
      "step": 80
    },
    {
      "epoch": 1.2686567164179103,
      "grad_norm": 2.0165800449063878,
      "learning_rate": 6.814152836799623e-07,
      "loss": 1.3597,
      "step": 85
    },
    {
      "epoch": 1.2686567164179103,
      "eval_loss": 1.6040061712265015,
      "eval_runtime": 5.9381,
      "eval_samples_per_second": 80.161,
      "eval_steps_per_second": 0.674,
      "step": 85
    },
    {
      "epoch": 1.3432835820895521,
      "grad_norm": 2.0046777857294593,
      "learning_rate": 5.178827427887253e-07,
      "loss": 1.3737,
      "step": 90
    },
    {
      "epoch": 1.3432835820895521,
      "eval_loss": 1.607108235359192,
      "eval_runtime": 5.938,
      "eval_samples_per_second": 80.161,
      "eval_steps_per_second": 0.674,
      "step": 90
    },
    {
      "epoch": 1.417910447761194,
      "grad_norm": 1.9599395887640554,
      "learning_rate": 3.938048900654646e-07,
      "loss": 1.3492,
      "step": 95
    },
    {
      "epoch": 1.417910447761194,
      "eval_loss": 1.607422113418579,
      "eval_runtime": 5.9586,
      "eval_samples_per_second": 79.884,
      "eval_steps_per_second": 0.671,
      "step": 95
    },
    {
      "epoch": 1.4925373134328357,
      "grad_norm": 2.090141489085601,
      "learning_rate": 3.0039745928104856e-07,
      "loss": 1.3826,
      "step": 100
    },
    {
      "epoch": 1.4925373134328357,
      "eval_loss": 1.6055359840393066,
      "eval_runtime": 5.9265,
      "eval_samples_per_second": 80.317,
      "eval_steps_per_second": 0.675,
      "step": 100
    },
    {
      "epoch": 1.5671641791044775,
      "grad_norm": 1.9332970593875183,
      "learning_rate": 2.3065986110942366e-07,
      "loss": 1.3533,
      "step": 105
    },
    {
      "epoch": 1.5671641791044775,
      "eval_loss": 1.6034619808197021,
      "eval_runtime": 5.9297,
      "eval_samples_per_second": 80.273,
      "eval_steps_per_second": 0.675,
      "step": 105
    },
    {
      "epoch": 1.6417910447761193,
      "grad_norm": 1.942043821463108,
      "learning_rate": 1.790498012203728e-07,
      "loss": 1.3611,
      "step": 110
    },
    {
      "epoch": 1.6417910447761193,
      "eval_loss": 1.6023228168487549,
      "eval_runtime": 5.9407,
      "eval_samples_per_second": 80.126,
      "eval_steps_per_second": 0.673,
      "step": 110
    },
    {
      "epoch": 1.716417910447761,
      "grad_norm": 1.9221561505361804,
      "learning_rate": 1.412101742620585e-07,
      "loss": 1.328,
      "step": 115
    },
    {
      "epoch": 1.716417910447761,
      "eval_loss": 1.602164387702942,
      "eval_runtime": 5.9392,
      "eval_samples_per_second": 80.145,
      "eval_steps_per_second": 0.673,
      "step": 115
    },
    {
      "epoch": 1.7910447761194028,
      "grad_norm": 1.8559815029587312,
      "learning_rate": 1.1374103686567517e-07,
      "loss": 1.3443,
      "step": 120
    },
    {
      "epoch": 1.7910447761194028,
      "eval_loss": 1.602574348449707,
      "eval_runtime": 5.9713,
      "eval_samples_per_second": 79.715,
      "eval_steps_per_second": 0.67,
      "step": 120
    },
    {
      "epoch": 1.8656716417910446,
      "grad_norm": 2.031347067054233,
      "learning_rate": 9.401027853897119e-08,
      "loss": 1.3386,
      "step": 125
    },
    {
      "epoch": 1.8656716417910446,
      "eval_loss": 1.602864146232605,
      "eval_runtime": 5.956,
      "eval_samples_per_second": 79.919,
      "eval_steps_per_second": 0.672,
      "step": 125
    },
    {
      "epoch": 1.9402985074626866,
      "grad_norm": 1.9772043505318835,
      "learning_rate": 7.999735362178058e-08,
      "loss": 1.3396,
      "step": 130
    },
    {
      "epoch": 1.9402985074626866,
      "eval_loss": 1.6029260158538818,
      "eval_runtime": 5.955,
      "eval_samples_per_second": 79.933,
      "eval_steps_per_second": 0.672,
      "step": 130
    },
    {
      "epoch": 2.014925373134328,
      "grad_norm": 2.0413931319541203,
      "learning_rate": 7.016511465398424e-08,
      "loss": 1.3573,
      "step": 135
    },
    {
      "epoch": 2.014925373134328,
      "eval_loss": 1.6028627157211304,
      "eval_runtime": 5.974,
      "eval_samples_per_second": 79.679,
      "eval_steps_per_second": 0.67,
      "step": 135
    },
    {
      "epoch": 2.08955223880597,
      "grad_norm": 1.9174789145055136,
      "learning_rate": 6.335540171983123e-08,
      "loss": 1.3754,
      "step": 140
    },
    {
      "epoch": 2.08955223880597,
      "eval_loss": 1.6034482717514038,
      "eval_runtime": 5.9455,
      "eval_samples_per_second": 80.061,
      "eval_steps_per_second": 0.673,
      "step": 140
    },
    {
      "epoch": 2.1641791044776117,
      "grad_norm": 1.8365955877771794,
      "learning_rate": 5.870459761414777e-08,
      "loss": 1.3229,
      "step": 145
    },
    {
      "epoch": 2.1641791044776117,
      "eval_loss": 1.6044023036956787,
      "eval_runtime": 5.9418,
      "eval_samples_per_second": 80.11,
      "eval_steps_per_second": 0.673,
      "step": 145
    },
    {
      "epoch": 2.2388059701492535,
      "grad_norm": 1.957661723231933,
      "learning_rate": 5.5575858924021805e-08,
      "loss": 1.3194,
      "step": 150
    },
    {
      "epoch": 2.2388059701492535,
      "eval_loss": 1.60551917552948,
      "eval_runtime": 5.9349,
      "eval_samples_per_second": 80.204,
      "eval_steps_per_second": 0.674,
      "step": 150
    },
    {
      "epoch": 2.3134328358208958,
      "grad_norm": 1.910367302653755,
      "learning_rate": 5.3505182098899865e-08,
      "loss": 1.3361,
      "step": 155
    },
    {
      "epoch": 2.3134328358208958,
      "eval_loss": 1.6064879894256592,
      "eval_runtime": 5.9543,
      "eval_samples_per_second": 79.942,
      "eval_steps_per_second": 0.672,
      "step": 155
    },
    {
      "epoch": 2.388059701492537,
      "grad_norm": 1.978379692085088,
      "learning_rate": 5.2158864924030916e-08,
      "loss": 1.3231,
      "step": 160
    },
    {
      "epoch": 2.388059701492537,
      "eval_loss": 1.60716712474823,
      "eval_runtime": 5.959,
      "eval_samples_per_second": 79.879,
      "eval_steps_per_second": 0.671,
      "step": 160
    },
    {
      "epoch": 2.4626865671641793,
      "grad_norm": 1.8179876847012644,
      "learning_rate": 5.130028101466846e-08,
      "loss": 1.32,
      "step": 165
    },
    {
      "epoch": 2.4626865671641793,
      "eval_loss": 1.6076022386550903,
      "eval_runtime": 5.918,
      "eval_samples_per_second": 80.433,
      "eval_steps_per_second": 0.676,
      "step": 165
    },
    {
      "epoch": 2.5373134328358207,
      "grad_norm": 1.947999010708536,
      "learning_rate": 5.076420137573488e-08,
      "loss": 1.3406,
      "step": 170
    },
    {
      "epoch": 2.5373134328358207,
      "eval_loss": 1.6078119277954102,
      "eval_runtime": 5.9402,
      "eval_samples_per_second": 80.132,
      "eval_steps_per_second": 0.673,
      "step": 170
    },
    {
      "epoch": 2.611940298507463,
      "grad_norm": 1.9301332893200993,
      "learning_rate": 5.0437175954586924e-08,
      "loss": 1.3184,
      "step": 175
    },
    {
      "epoch": 2.611940298507463,
      "eval_loss": 1.6078969240188599,
      "eval_runtime": 5.9794,
      "eval_samples_per_second": 79.607,
      "eval_steps_per_second": 0.669,
      "step": 175
    },
    {
      "epoch": 2.6865671641791042,
      "grad_norm": 1.8025714970137283,
      "learning_rate": 5.024273255042209e-08,
      "loss": 1.2745,
      "step": 180
    },
    {
      "epoch": 2.6865671641791042,
      "eval_loss": 1.607950210571289,
      "eval_runtime": 5.9775,
      "eval_samples_per_second": 79.632,
      "eval_steps_per_second": 0.669,
      "step": 180
    },
    {
      "epoch": 2.7611940298507465,
      "grad_norm": 1.8246857333350315,
      "learning_rate": 5.013036339654868e-08,
      "loss": 1.3024,
      "step": 185
    },
    {
      "epoch": 2.7611940298507465,
      "eval_loss": 1.607932448387146,
      "eval_runtime": 5.9386,
      "eval_samples_per_second": 80.154,
      "eval_steps_per_second": 0.674,
      "step": 185
    },
    {
      "epoch": 2.835820895522388,
      "grad_norm": 1.8544541968240376,
      "learning_rate": 5.006745403152126e-08,
      "loss": 1.3243,
      "step": 190
    },
    {
      "epoch": 2.835820895522388,
      "eval_loss": 1.6079127788543701,
      "eval_runtime": 5.9362,
      "eval_samples_per_second": 80.186,
      "eval_steps_per_second": 0.674,
      "step": 190
    },
    {
      "epoch": 2.91044776119403,
      "grad_norm": 1.9345137440066773,
      "learning_rate": 5.0033467418823265e-08,
      "loss": 1.3239,
      "step": 195
    },
    {
      "epoch": 2.91044776119403,
      "eval_loss": 1.6080163717269897,
      "eval_runtime": 5.9581,
      "eval_samples_per_second": 79.892,
      "eval_steps_per_second": 0.671,
      "step": 195
    },
    {
      "epoch": 2.9850746268656714,
      "grad_norm": 2.033807758057437,
      "learning_rate": 5.001583122566221e-08,
      "loss": 1.3349,
      "step": 200
    },
    {
      "epoch": 2.9850746268656714,
      "eval_loss": 1.6080641746520996,
      "eval_runtime": 5.9287,
      "eval_samples_per_second": 80.287,
      "eval_steps_per_second": 0.675,
      "step": 200
    },
    {
      "epoch": 3.0597014925373136,
      "grad_norm": 1.923714106710108,
      "learning_rate": 5.0007090159326376e-08,
      "loss": 1.337,
      "step": 205
    },
    {
      "epoch": 3.0597014925373136,
      "eval_loss": 1.6079014539718628,
      "eval_runtime": 5.9355,
      "eval_samples_per_second": 80.195,
      "eval_steps_per_second": 0.674,
      "step": 205
    },
    {
      "epoch": 3.1343283582089554,
      "grad_norm": 1.8059505650037335,
      "learning_rate": 5.000298058072559e-08,
      "loss": 1.3091,
      "step": 210
    },
    {
      "epoch": 3.1343283582089554,
      "eval_loss": 1.6077641248703003,
      "eval_runtime": 5.9641,
      "eval_samples_per_second": 79.811,
      "eval_steps_per_second": 0.671,
      "step": 210
    },
    {
      "epoch": 3.208955223880597,
      "grad_norm": 1.894523950936326,
      "learning_rate": 5.0001163431988796e-08,
      "loss": 1.3266,
      "step": 215
    },
    {
      "epoch": 3.208955223880597,
      "eval_loss": 1.60787832736969,
      "eval_runtime": 5.9618,
      "eval_samples_per_second": 79.841,
      "eval_steps_per_second": 0.671,
      "step": 215
    },
    {
      "epoch": 3.283582089552239,
      "grad_norm": 1.8919183390013388,
      "learning_rate": 5.0000415857597904e-08,
      "loss": 1.3014,
      "step": 220
    },
    {
      "epoch": 3.283582089552239,
      "eval_loss": 1.6083022356033325,
      "eval_runtime": 5.9802,
      "eval_samples_per_second": 79.596,
      "eval_steps_per_second": 0.669,
      "step": 220
    },
    {
      "epoch": 3.3582089552238807,
      "grad_norm": 1.8867080087928592,
      "learning_rate": 5.0000133662276014e-08,
      "loss": 1.3153,
      "step": 225
    },
    {
      "epoch": 3.3582089552238807,
      "eval_loss": 1.608628273010254,
      "eval_runtime": 5.9587,
      "eval_samples_per_second": 79.883,
      "eval_steps_per_second": 0.671,
      "step": 225
    },
    {
      "epoch": 3.4328358208955225,
      "grad_norm": 2.000197758434265,
      "learning_rate": 5.0000037695987735e-08,
      "loss": 1.3192,
      "step": 230
    },
    {
      "epoch": 3.4328358208955225,
      "eval_loss": 1.6089787483215332,
      "eval_runtime": 5.931,
      "eval_samples_per_second": 80.256,
      "eval_steps_per_second": 0.674,
      "step": 230
    },
    {
      "epoch": 3.5074626865671643,
      "grad_norm": 1.8236033935396483,
      "learning_rate": 5.0000009015715774e-08,
      "loss": 1.315,
      "step": 235
    },
    {
      "epoch": 3.5074626865671643,
      "eval_loss": 1.6092561483383179,
      "eval_runtime": 5.9999,
      "eval_samples_per_second": 79.335,
      "eval_steps_per_second": 0.667,
      "step": 235
    },
    {
      "epoch": 3.582089552238806,
      "grad_norm": 1.9175089422860454,
      "learning_rate": 5.0000001740409057e-08,
      "loss": 1.3047,
      "step": 240
    },
    {
      "epoch": 3.582089552238806,
      "eval_loss": 1.6093214750289917,
      "eval_runtime": 5.9889,
      "eval_samples_per_second": 79.481,
      "eval_steps_per_second": 0.668,
      "step": 240
    },
    {
      "epoch": 3.656716417910448,
      "grad_norm": 1.8702465201892255,
      "learning_rate": 5.000000025140825e-08,
      "loss": 1.3208,
      "step": 245
    },
    {
      "epoch": 3.656716417910448,
      "eval_loss": 1.609329342842102,
      "eval_runtime": 5.9315,
      "eval_samples_per_second": 80.25,
      "eval_steps_per_second": 0.674,
      "step": 245
    },
    {
      "epoch": 3.7313432835820897,
      "grad_norm": 1.9567098448807434,
      "learning_rate": 5.000000002399302e-08,
      "loss": 1.362,
      "step": 250
    },
    {
      "epoch": 3.7313432835820897,
      "eval_loss": 1.609269380569458,
      "eval_runtime": 5.9582,
      "eval_samples_per_second": 79.89,
      "eval_steps_per_second": 0.671,
      "step": 250
    },
    {
      "epoch": 3.8059701492537314,
      "grad_norm": 1.8742551595721921,
      "learning_rate": 5.0000000001205015e-08,
      "loss": 1.3255,
      "step": 255
    },
    {
      "epoch": 3.8059701492537314,
      "eval_loss": 1.609143853187561,
      "eval_runtime": 5.9335,
      "eval_samples_per_second": 80.223,
      "eval_steps_per_second": 0.674,
      "step": 255
    },
    {
      "epoch": 3.8805970149253732,
      "grad_norm": 1.9100964620707728,
      "learning_rate": 5.000000000001954e-08,
      "loss": 1.2941,
      "step": 260
    },
    {
      "epoch": 3.8805970149253732,
      "eval_loss": 1.608902096748352,
      "eval_runtime": 5.9228,
      "eval_samples_per_second": 80.368,
      "eval_steps_per_second": 0.675,
      "step": 260
    },
    {
      "epoch": 3.955223880597015,
      "grad_norm": 1.959684298255238,
      "learning_rate": 5.0000000000000024e-08,
      "loss": 1.3254,
      "step": 265
    },
    {
      "epoch": 3.955223880597015,
      "eval_loss": 1.6086210012435913,
      "eval_runtime": 5.9667,
      "eval_samples_per_second": 79.776,
      "eval_steps_per_second": 0.67,
      "step": 265
    },
    {
      "epoch": 4.0,
      "step": 268,
      "total_flos": 27742804377600.0,
      "train_loss": 1.416380336925165,
      "train_runtime": 8905.9803,
      "train_samples_per_second": 1.923,
      "train_steps_per_second": 0.03
    }
  ],
  "logging_steps": 5,
  "max_steps": 268,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 5,
  "total_flos": 27742804377600.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}