{
  "best_global_step": 360,
  "best_metric": 0.19541434943675995,
  "best_model_checkpoint": "./llava_adalora_weather_model/checkpoint-360",
  "epoch": 3.67866323907455,
  "eval_steps": 20,
  "global_step": 360,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.10282776349614396,
      "grad_norm": 0.5032290816307068,
      "learning_rate": 6.75e-05,
      "loss": 2.3108,
      "mean_token_accuracy": 0.5181779790669679,
      "num_tokens": 645254.0,
      "step": 10
    },
    {
      "epoch": 0.20565552699228792,
      "grad_norm": 0.5140016078948975,
      "learning_rate": 0.0001425,
      "loss": 1.9876,
      "mean_token_accuracy": 0.5564090937376023,
      "num_tokens": 1290816.0,
      "step": 20
    },
    {
      "epoch": 0.20565552699228792,
      "eval_loss": 1.6786144971847534,
      "eval_mean_token_accuracy": 0.5918497213950524,
      "eval_num_tokens": 1290816.0,
      "eval_runtime": 199.77,
      "eval_samples_per_second": 0.976,
      "eval_steps_per_second": 0.976,
      "step": 20
    },
    {
      "epoch": 0.30848329048843187,
      "grad_norm": 1.200245976448059,
      "learning_rate": 0.00021749999999999997,
      "loss": 1.2898,
      "mean_token_accuracy": 0.6574281774461269,
      "num_tokens": 1936067.0,
      "step": 30
    },
    {
      "epoch": 0.41131105398457585,
      "grad_norm": 0.3560621440410614,
      "learning_rate": 0.00029249999999999995,
      "loss": 0.5269,
      "mean_token_accuracy": 0.8504049643874169,
      "num_tokens": 2581570.0,
      "step": 40
    },
    {
      "epoch": 0.41131105398457585,
      "eval_loss": 0.3815372884273529,
      "eval_mean_token_accuracy": 0.8895098340816987,
      "eval_num_tokens": 2581570.0,
      "eval_runtime": 199.91,
      "eval_samples_per_second": 0.975,
      "eval_steps_per_second": 0.975,
      "step": 40
    },
    {
      "epoch": 0.5141388174807198,
      "grad_norm": 0.36290714144706726,
      "learning_rate": 0.0002995163544683256,
      "loss": 0.3431,
      "mean_token_accuracy": 0.8966262958943844,
      "num_tokens": 3226740.0,
      "step": 50
    },
    {
      "epoch": 0.6169665809768637,
      "grad_norm": 0.3132888674736023,
      "learning_rate": 0.00029784849709745616,
      "loss": 0.3131,
      "mean_token_accuracy": 0.9030718393623829,
      "num_tokens": 3872000.0,
      "step": 60
    },
    {
      "epoch": 0.6169665809768637,
      "eval_loss": 0.2877984046936035,
      "eval_mean_token_accuracy": 0.9094783853261899,
      "eval_num_tokens": 3872000.0,
      "eval_runtime": 200.0291,
      "eval_samples_per_second": 0.975,
      "eval_steps_per_second": 0.975,
      "step": 60
    },
    {
      "epoch": 0.7197943444730077,
      "grad_norm": 0.3172520697116852,
      "learning_rate": 0.0002950037303267096,
      "loss": 0.2776,
      "mean_token_accuracy": 0.9121568284928798,
      "num_tokens": 4517650.0,
      "step": 70
    },
    {
      "epoch": 0.8226221079691517,
      "grad_norm": 0.2681174576282501,
      "learning_rate": 0.0002910046991800035,
      "loss": 0.2463,
      "mean_token_accuracy": 0.9197294652462006,
      "num_tokens": 5162524.0,
      "step": 80
    },
    {
      "epoch": 0.8226221079691517,
      "eval_loss": 0.24699676036834717,
      "eval_mean_token_accuracy": 0.918833449559334,
      "eval_num_tokens": 5162524.0,
      "eval_runtime": 199.9764,
      "eval_samples_per_second": 0.975,
      "eval_steps_per_second": 0.975,
      "step": 80
    },
    {
      "epoch": 0.9254498714652957,
      "grad_norm": 0.25532031059265137,
      "learning_rate": 0.00028588323690176954,
      "loss": 0.2409,
      "mean_token_accuracy": 0.9201722621917725,
      "num_tokens": 5808159.0,
      "step": 90
    },
    {
      "epoch": 1.0205655526992288,
      "grad_norm": 0.2883255183696747,
      "learning_rate": 0.0002796801115567139,
      "loss": 0.229,
      "mean_token_accuracy": 0.9240316643908217,
      "num_tokens": 6405024.0,
      "step": 100
    },
    {
      "epoch": 1.0205655526992288,
      "eval_loss": 0.2340758889913559,
      "eval_mean_token_accuracy": 0.9224844058354695,
      "eval_num_tokens": 6405024.0,
      "eval_runtime": 199.917,
      "eval_samples_per_second": 0.975,
      "eval_steps_per_second": 0.975,
      "step": 100
    },
    {
      "epoch": 1.1233933161953726,
      "grad_norm": 0.30399489402770996,
      "learning_rate": 0.0002724447015062708,
      "loss": 0.2179,
      "mean_token_accuracy": 0.927613090723753,
      "num_tokens": 7050233.0,
      "step": 110
    },
    {
      "epoch": 1.2262210796915167,
      "grad_norm": 0.28451991081237793,
      "learning_rate": 0.0002642346023450357,
      "loss": 0.216,
      "mean_token_accuracy": 0.9265601448714733,
      "num_tokens": 7695653.0,
      "step": 120
    },
    {
      "epoch": 1.2262210796915167,
      "eval_loss": 0.2247939258813858,
      "eval_mean_token_accuracy": 0.925137395125169,
      "eval_num_tokens": 7695653.0,
      "eval_runtime": 199.5418,
      "eval_samples_per_second": 0.977,
      "eval_steps_per_second": 0.977,
      "step": 120
    },
    {
      "epoch": 1.3290488431876606,
      "grad_norm": 0.28032347559928894,
      "learning_rate": 0.0002551151684260553,
      "loss": 0.2093,
      "mean_token_accuracy": 0.9289478570222854,
      "num_tokens": 8341317.0,
      "step": 130
    },
    {
      "epoch": 1.4318766066838047,
      "grad_norm": 0.3006415367126465,
      "learning_rate": 0.0002451589926245468,
      "loss": 0.2017,
      "mean_token_accuracy": 0.930090955644846,
      "num_tokens": 8986431.0,
      "step": 140
    },
    {
      "epoch": 1.4318766066838047,
      "eval_loss": 0.22003041207790375,
      "eval_mean_token_accuracy": 0.9265705820841666,
      "eval_num_tokens": 8986431.0,
      "eval_runtime": 199.7226,
      "eval_samples_per_second": 0.976,
      "eval_steps_per_second": 0.976,
      "step": 140
    },
    {
      "epoch": 1.5347043701799485,
      "grad_norm": 0.27291589975357056,
      "learning_rate": 0.00023444532848124715,
      "loss": 0.2008,
      "mean_token_accuracy": 0.9297313310205937,
      "num_tokens": 9631882.0,
      "step": 150
    },
    {
      "epoch": 1.6375321336760926,
      "grad_norm": 0.2605527639389038,
      "learning_rate": 0.00022305945932527308,
      "loss": 0.2019,
      "mean_token_accuracy": 0.9302895963191986,
      "num_tokens": 10277533.0,
      "step": 160
    },
    {
      "epoch": 1.6375321336760926,
      "eval_loss": 0.21295307576656342,
      "eval_mean_token_accuracy": 0.9283858265632238,
      "eval_num_tokens": 10277533.0,
      "eval_runtime": 199.8041,
      "eval_samples_per_second": 0.976,
      "eval_steps_per_second": 0.976,
      "step": 160
    },
    {
      "epoch": 1.7403598971722365,
      "grad_norm": 0.2581302523612976,
      "learning_rate": 0.0002110920193984228,
      "loss": 0.2082,
      "mean_token_accuracy": 0.9276637695729733,
      "num_tokens": 10922895.0,
      "step": 170
    },
    {
      "epoch": 1.8431876606683804,
      "grad_norm": 0.2544514238834381,
      "learning_rate": 0.00019863827238493308,
      "loss": 0.1954,
      "mean_token_accuracy": 0.9318794839084148,
      "num_tokens": 11568708.0,
      "step": 180
    },
    {
      "epoch": 1.8431876606683804,
      "eval_loss": 0.20810775458812714,
      "eval_mean_token_accuracy": 0.9295981232936565,
      "eval_num_tokens": 11568708.0,
      "eval_runtime": 200.1655,
      "eval_samples_per_second": 0.974,
      "eval_steps_per_second": 0.974,
      "step": 180
    },
    {
      "epoch": 1.9460154241645244,
      "grad_norm": 0.2642477750778198,
      "learning_rate": 0.00018579735308976727,
      "loss": 0.1966,
      "mean_token_accuracy": 0.9328551657497883,
      "num_tokens": 12213156.0,
      "step": 190
    },
    {
      "epoch": 2.0411311053984575,
      "grad_norm": 0.25700676441192627,
      "learning_rate": 0.00017267147830185608,
      "loss": 0.1901,
      "mean_token_accuracy": 0.9336085150370727,
      "num_tokens": 12809961.0,
      "step": 200
    },
    {
      "epoch": 2.0411311053984575,
      "eval_loss": 0.20741970837116241,
      "eval_mean_token_accuracy": 0.9298222092481759,
      "eval_num_tokens": 12809961.0,
      "eval_runtime": 199.9295,
      "eval_samples_per_second": 0.975,
      "eval_steps_per_second": 0.975,
      "step": 200
    },
    {
      "epoch": 2.1439588688946016,
      "grad_norm": 0.27311551570892334,
      "learning_rate": 0.00015936513312400936,
      "loss": 0.1743,
      "mean_token_accuracy": 0.9388107411563397,
      "num_tokens": 13455171.0,
      "step": 210
    },
    {
      "epoch": 2.2467866323907453,
      "grad_norm": 0.24801801145076752,
      "learning_rate": 0.0001459842392465063,
      "loss": 0.1771,
      "mean_token_accuracy": 0.9377820797264576,
      "num_tokens": 14099963.0,
      "step": 220
    },
    {
      "epoch": 2.2467866323907453,
      "eval_loss": 0.2031983733177185,
      "eval_mean_token_accuracy": 0.931696018194541,
      "eval_num_tokens": 14099963.0,
      "eval_runtime": 199.9633,
      "eval_samples_per_second": 0.975,
      "eval_steps_per_second": 0.975,
      "step": 220
    },
    {
      "epoch": 2.3496143958868894,
      "grad_norm": 0.25646060705184937,
      "learning_rate": 0.00013263531178510647,
      "loss": 0.1704,
      "mean_token_accuracy": 0.9393708236515522,
      "num_tokens": 14745386.0,
      "step": 230
    },
    {
      "epoch": 2.4524421593830334,
      "grad_norm": 0.27255287766456604,
      "learning_rate": 0.00011942461139525123,
      "loss": 0.1755,
      "mean_token_accuracy": 0.9370259158313274,
      "num_tokens": 15390859.0,
      "step": 240
    },
    {
      "epoch": 2.4524421593830334,
      "eval_loss": 0.2018389254808426,
      "eval_mean_token_accuracy": 0.9325766147711338,
      "eval_num_tokens": 15390859.0,
      "eval_runtime": 199.4447,
      "eval_samples_per_second": 0.978,
      "eval_steps_per_second": 0.978,
      "step": 240
    },
    {
      "epoch": 2.5552699228791775,
      "grad_norm": 0.2926770746707916,
      "learning_rate": 0.00010645729841183066,
      "loss": 0.1752,
      "mean_token_accuracy": 0.9387852221727371,
      "num_tokens": 16035959.0,
      "step": 250
    },
    {
      "epoch": 2.658097686375321,
      "grad_norm": 0.2695241868495941,
      "learning_rate": 9.383659574776544e-05,
      "loss": 0.1722,
      "mean_token_accuracy": 0.9387590140104294,
      "num_tokens": 16681649.0,
      "step": 260
    },
    {
      "epoch": 2.658097686375321,
      "eval_loss": 0.1990961730480194,
      "eval_mean_token_accuracy": 0.9332608406360333,
      "eval_num_tokens": 16681649.0,
      "eval_runtime": 199.8342,
      "eval_samples_per_second": 0.976,
      "eval_steps_per_second": 0.976,
      "step": 260
    },
    {
      "epoch": 2.7609254498714653,
      "grad_norm": 0.263235479593277,
      "learning_rate": 8.166296721493059e-05,
      "loss": 0.1752,
      "mean_token_accuracy": 0.9383706137537956,
      "num_tokens": 17327083.0,
      "step": 270
    },
    {
      "epoch": 2.8637532133676094,
      "grad_norm": 0.28708621859550476,
      "learning_rate": 7.003331780818343e-05,
      "loss": 0.1639,
      "mean_token_accuracy": 0.9409998580813408,
      "num_tokens": 17972810.0,
      "step": 280
    },
    {
      "epoch": 2.8637532133676094,
      "eval_loss": 0.19682233035564423,
      "eval_mean_token_accuracy": 0.9337082728361472,
      "eval_num_tokens": 17972810.0,
      "eval_runtime": 200.3343,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 280
    },
    {
      "epoch": 2.966580976863753,
      "grad_norm": 0.27147579193115234,
      "learning_rate": 5.9040222318426706e-05,
      "loss": 0.1665,
      "mean_token_accuracy": 0.9407577112317085,
      "num_tokens": 18618053.0,
      "step": 290
    },
    {
      "epoch": 3.0616966580976865,
      "grad_norm": 0.264649897813797,
      "learning_rate": 4.8771188415130426e-05,
      "loss": 0.1615,
      "mean_token_accuracy": 0.9429288788421734,
      "num_tokens": 19215133.0,
      "step": 300
    },
    {
      "epoch": 3.0616966580976865,
      "eval_loss": 0.19712290167808533,
      "eval_mean_token_accuracy": 0.9342636979543246,
      "eval_num_tokens": 19215133.0,
      "eval_runtime": 200.8771,
      "eval_samples_per_second": 0.971,
      "eval_steps_per_second": 0.971,
      "step": 300
    },
    {
      "epoch": 3.16452442159383,
      "grad_norm": 0.2663537263870239,
      "learning_rate": 3.930796006435253e-05,
      "loss": 0.1575,
      "mean_token_accuracy": 0.9434727318584919,
      "num_tokens": 19860293.0,
      "step": 310
    },
    {
      "epoch": 3.2673521850899743,
      "grad_norm": 0.2932131290435791,
      "learning_rate": 3.072586682721707e-05,
      "loss": 0.15,
      "mean_token_accuracy": 0.9469764873385429,
      "num_tokens": 20505458.0,
      "step": 320
    },
    {
      "epoch": 3.2673521850899743,
      "eval_loss": 0.19779063761234283,
      "eval_mean_token_accuracy": 0.9343466890163911,
      "eval_num_tokens": 20505458.0,
      "eval_runtime": 200.8092,
      "eval_samples_per_second": 0.971,
      "eval_steps_per_second": 0.971,
      "step": 320
    },
    {
      "epoch": 3.3701799485861184,
      "grad_norm": 0.270320326089859,
      "learning_rate": 2.3093224218590266e-05,
      "loss": 0.152,
      "mean_token_accuracy": 0.9450157150626183,
      "num_tokens": 21150733.0,
      "step": 330
    },
    {
      "epoch": 3.4730077120822624,
      "grad_norm": 0.27023226022720337,
      "learning_rate": 1.6470789899242098e-05,
      "loss": 0.1573,
      "mean_token_accuracy": 0.9442848064005375,
      "num_tokens": 21795902.0,
      "step": 340
    },
    {
      "epoch": 3.4730077120822624,
      "eval_loss": 0.19579839706420898,
      "eval_mean_token_accuracy": 0.9347658371314024,
      "eval_num_tokens": 21795902.0,
      "eval_runtime": 200.3793,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 340
    },
    {
      "epoch": 3.575835475578406,
      "grad_norm": 0.2818222939968109,
      "learning_rate": 1.0911280030334307e-05,
      "loss": 0.1527,
      "mean_token_accuracy": 0.9446957901120185,
      "num_tokens": 22441445.0,
      "step": 350
    },
    {
      "epoch": 3.67866323907455,
      "grad_norm": 0.28465282917022705,
      "learning_rate": 6.458949640168675e-06,
      "loss": 0.1499,
      "mean_token_accuracy": 0.9464483924210072,
      "num_tokens": 23087068.0,
      "step": 360
    },
    {
      "epoch": 3.67866323907455,
      "eval_loss": 0.19541434943675995,
      "eval_mean_token_accuracy": 0.9350230287282895,
      "eval_num_tokens": 23087068.0,
      "eval_runtime": 200.3568,
      "eval_samples_per_second": 0.973,
      "eval_steps_per_second": 0.973,
      "step": 360
    }
  ],
  "logging_steps": 10,
  "max_steps": 392,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 40,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.038151612191998e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}