{
  "best_metric": 0.3039023280143738,
  "best_model_checkpoint": "./kumpun-output/checkpoint-400",
  "epoch": 2.1164021164021163,
  "eval_steps": 200,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "grad_norm": 5.246432304382324,
      "learning_rate": 7.000000000000001e-05,
      "loss": 4.0922,
      "step": 10
    },
    {
      "epoch": 0.11,
      "grad_norm": 3.076571226119995,
      "learning_rate": 0.00017,
      "loss": 3.5074,
      "step": 20
    },
    {
      "epoch": 0.16,
      "grad_norm": 11.657795906066895,
      "learning_rate": 0.00026000000000000003,
      "loss": 2.2509,
      "step": 30
    },
    {
      "epoch": 0.21,
      "grad_norm": 3.2525696754455566,
      "learning_rate": 0.00034,
      "loss": 0.8157,
      "step": 40
    },
    {
      "epoch": 0.26,
      "grad_norm": 2.755035400390625,
      "learning_rate": 0.00044,
      "loss": 0.5126,
      "step": 50
    },
    {
      "epoch": 0.32,
      "grad_norm": 2.1165263652801514,
      "learning_rate": 0.00054,
      "loss": 0.4988,
      "step": 60
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.805057168006897,
      "learning_rate": 0.00064,
      "loss": 0.2959,
      "step": 70
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.8083710074424744,
      "learning_rate": 0.00074,
      "loss": 0.2779,
      "step": 80
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.7004362344741821,
      "learning_rate": 0.00084,
      "loss": 0.244,
      "step": 90
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.5553646087646484,
      "learning_rate": 0.00094,
      "loss": 0.3729,
      "step": 100
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.8424310684204102,
      "learning_rate": 0.0009914346895074947,
      "loss": 0.3408,
      "step": 110
    },
    {
      "epoch": 0.63,
      "grad_norm": 6.2793192863464355,
      "learning_rate": 0.0009700214132762313,
      "loss": 0.3011,
      "step": 120
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.8749362826347351,
      "learning_rate": 0.0009486081370449678,
      "loss": 0.249,
      "step": 130
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.6772909760475159,
      "learning_rate": 0.0009271948608137045,
      "loss": 0.2258,
      "step": 140
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.4329707622528076,
      "learning_rate": 0.0009057815845824411,
      "loss": 0.2977,
      "step": 150
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.501881718635559,
      "learning_rate": 0.0008843683083511778,
      "loss": 0.2276,
      "step": 160
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.2264193296432495,
      "learning_rate": 0.0008629550321199144,
      "loss": 0.2724,
      "step": 170
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.8799680471420288,
      "learning_rate": 0.0008415417558886511,
      "loss": 0.2358,
      "step": 180
    },
    {
      "epoch": 1.01,
      "grad_norm": 23.913034439086914,
      "learning_rate": 0.0008201284796573875,
      "loss": 0.4151,
      "step": 190
    },
    {
      "epoch": 1.06,
      "grad_norm": 2.3118550777435303,
      "learning_rate": 0.0007987152034261242,
      "loss": 0.4287,
      "step": 200
    },
    {
      "epoch": 1.06,
      "eval_loss": 0.3133811950683594,
      "eval_runtime": 0.6704,
      "eval_samples_per_second": 14.916,
      "eval_steps_per_second": 2.983,
      "step": 200
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.6871071457862854,
      "learning_rate": 0.0007773019271948608,
      "loss": 0.2392,
      "step": 210
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.5935308933258057,
      "learning_rate": 0.0007558886509635975,
      "loss": 0.225,
      "step": 220
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.6274584531784058,
      "learning_rate": 0.0007344753747323341,
      "loss": 0.174,
      "step": 230
    },
    {
      "epoch": 1.27,
      "grad_norm": 1.8129804134368896,
      "learning_rate": 0.0007130620985010708,
      "loss": 0.2524,
      "step": 240
    },
    {
      "epoch": 1.32,
      "grad_norm": 1.9421052932739258,
      "learning_rate": 0.0006916488222698073,
      "loss": 0.2796,
      "step": 250
    },
    {
      "epoch": 1.38,
      "grad_norm": 1.4675747156143188,
      "learning_rate": 0.0006702355460385438,
      "loss": 0.25,
      "step": 260
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.9965176582336426,
      "learning_rate": 0.0006488222698072805,
      "loss": 0.212,
      "step": 270
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.8673143982887268,
      "learning_rate": 0.0006274089935760171,
      "loss": 0.1898,
      "step": 280
    },
    {
      "epoch": 1.53,
      "grad_norm": 2.1284236907958984,
      "learning_rate": 0.0006059957173447538,
      "loss": 0.3798,
      "step": 290
    },
    {
      "epoch": 1.59,
      "grad_norm": 2.246885299682617,
      "learning_rate": 0.0005845824411134904,
      "loss": 0.2751,
      "step": 300
    },
    {
      "epoch": 1.64,
      "grad_norm": 1.626805305480957,
      "learning_rate": 0.000563169164882227,
      "loss": 0.2648,
      "step": 310
    },
    {
      "epoch": 1.69,
      "grad_norm": 1.273806095123291,
      "learning_rate": 0.0005417558886509636,
      "loss": 0.2163,
      "step": 320
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.5863224267959595,
      "learning_rate": 0.0005203426124197003,
      "loss": 0.1584,
      "step": 330
    },
    {
      "epoch": 1.8,
      "grad_norm": 1.6265116930007935,
      "learning_rate": 0.0004989293361884369,
      "loss": 0.3904,
      "step": 340
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.6448193192481995,
      "learning_rate": 0.00047751605995717344,
      "loss": 0.2429,
      "step": 350
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.6662893295288086,
      "learning_rate": 0.0004561027837259101,
      "loss": 0.1899,
      "step": 360
    },
    {
      "epoch": 1.96,
      "grad_norm": 0.8907824158668518,
      "learning_rate": 0.0004346895074946467,
      "loss": 0.2035,
      "step": 370
    },
    {
      "epoch": 2.01,
      "grad_norm": 1.2576186656951904,
      "learning_rate": 0.0004132762312633833,
      "loss": 0.2224,
      "step": 380
    },
    {
      "epoch": 2.06,
      "grad_norm": 1.1375278234481812,
      "learning_rate": 0.00039186295503211994,
      "loss": 0.2645,
      "step": 390
    },
    {
      "epoch": 2.12,
      "grad_norm": 1.2520960569381714,
      "learning_rate": 0.0003704496788008566,
      "loss": 0.2,
      "step": 400
    },
    {
      "epoch": 2.12,
      "eval_loss": 0.3039023280143738,
      "eval_runtime": 0.6751,
      "eval_samples_per_second": 14.814,
      "eval_steps_per_second": 2.963,
      "step": 400
    }
  ],
  "logging_steps": 10,
  "max_steps": 567,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 200,
  "total_flos": 2456831348441088.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}