{
  "best_global_step": 900,
  "best_metric": 0.8113172650337219,
  "best_model_checkpoint": "/home/jovyan/work/pad3/hf_cache/vlm-ministral_cpt/checkpoint-900",
  "epoch": 1.0285714285714285,
  "eval_steps": 300,
  "global_step": 900,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.011428571428571429,
      "grad_norm": 15.378578186035156,
      "learning_rate": 3.39622641509434e-05,
      "loss": 8.33862533569336,
      "step": 10
    },
    {
      "epoch": 0.022857142857142857,
      "grad_norm": 10.927111625671387,
      "learning_rate": 7.169811320754717e-05,
      "loss": 6.778567504882813,
      "step": 20
    },
    {
      "epoch": 0.03428571428571429,
      "grad_norm": 9.256351470947266,
      "learning_rate": 0.00010943396226415095,
      "loss": 5.987285232543945,
      "step": 30
    },
    {
      "epoch": 0.045714285714285714,
      "grad_norm": 8.331721305847168,
      "learning_rate": 0.00014716981132075472,
      "loss": 5.4672698974609375,
      "step": 40
    },
    {
      "epoch": 0.05714285714285714,
      "grad_norm": 8.703407287597656,
      "learning_rate": 0.0001849056603773585,
      "loss": 5.27801742553711,
      "step": 50
    },
    {
      "epoch": 0.06857142857142857,
      "grad_norm": 9.705166816711426,
      "learning_rate": 0.0001992928697701827,
      "loss": 5.180379486083984,
      "step": 60
    },
    {
      "epoch": 0.08,
      "grad_norm": 9.159878730773926,
      "learning_rate": 0.00019811431938715382,
      "loss": 5.042933654785156,
      "step": 70
    },
    {
      "epoch": 0.09142857142857143,
      "grad_norm": 7.970294952392578,
      "learning_rate": 0.00019693576900412494,
      "loss": 4.798321151733399,
      "step": 80
    },
    {
      "epoch": 0.10285714285714286,
      "grad_norm": 8.137990951538086,
      "learning_rate": 0.00019575721862109607,
      "loss": 4.7907672882080075,
      "step": 90
    },
    {
      "epoch": 0.11428571428571428,
      "grad_norm": 7.5624799728393555,
      "learning_rate": 0.0001945786682380672,
      "loss": 4.674889755249024,
      "step": 100
    },
    {
      "epoch": 0.12571428571428572,
      "grad_norm": 7.773468017578125,
      "learning_rate": 0.00019340011785503832,
      "loss": 4.581521606445312,
      "step": 110
    },
    {
      "epoch": 0.13714285714285715,
      "grad_norm": 7.638908386230469,
      "learning_rate": 0.00019222156747200942,
      "loss": 4.639577484130859,
      "step": 120
    },
    {
      "epoch": 0.14857142857142858,
      "grad_norm": 8.217121124267578,
      "learning_rate": 0.00019104301708898055,
      "loss": 4.598583221435547,
      "step": 130
    },
    {
      "epoch": 0.16,
      "grad_norm": 7.484283447265625,
      "learning_rate": 0.00018986446670595167,
      "loss": 4.600800704956055,
      "step": 140
    },
    {
      "epoch": 0.17142857142857143,
      "grad_norm": 7.393186569213867,
      "learning_rate": 0.0001886859163229228,
      "loss": 4.368617248535156,
      "step": 150
    },
    {
      "epoch": 0.18285714285714286,
      "grad_norm": 7.616724014282227,
      "learning_rate": 0.00018750736593989393,
      "loss": 4.389640808105469,
      "step": 160
    },
    {
      "epoch": 0.19428571428571428,
      "grad_norm": 7.035678863525391,
      "learning_rate": 0.00018632881555686505,
      "loss": 4.333992385864258,
      "step": 170
    },
    {
      "epoch": 0.2057142857142857,
      "grad_norm": 6.236395359039307,
      "learning_rate": 0.00018515026517383618,
      "loss": 4.294047164916992,
      "step": 180
    },
    {
      "epoch": 0.21714285714285714,
      "grad_norm": 6.82771110534668,
      "learning_rate": 0.00018397171479080733,
      "loss": 4.332611465454102,
      "step": 190
    },
    {
      "epoch": 0.22857142857142856,
      "grad_norm": 6.608238697052002,
      "learning_rate": 0.00018279316440777846,
      "loss": 4.1796875,
      "step": 200
    },
    {
      "epoch": 0.24,
      "grad_norm": 6.314072132110596,
      "learning_rate": 0.00018161461402474958,
      "loss": 4.134822845458984,
      "step": 210
    },
    {
      "epoch": 0.25142857142857145,
      "grad_norm": 6.577261924743652,
      "learning_rate": 0.0001804360636417207,
      "loss": 4.200227355957031,
      "step": 220
    },
    {
      "epoch": 0.26285714285714284,
      "grad_norm": 5.533579349517822,
      "learning_rate": 0.00017925751325869184,
      "loss": 4.231822204589844,
      "step": 230
    },
    {
      "epoch": 0.2742857142857143,
      "grad_norm": 5.7535552978515625,
      "learning_rate": 0.00017807896287566294,
      "loss": 4.185052871704102,
      "step": 240
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 6.131353855133057,
      "learning_rate": 0.00017690041249263406,
      "loss": 4.047256851196289,
      "step": 250
    },
    {
      "epoch": 0.29714285714285715,
      "grad_norm": 5.759047508239746,
      "learning_rate": 0.0001757218621096052,
      "loss": 4.180568695068359,
      "step": 260
    },
    {
      "epoch": 0.30857142857142855,
      "grad_norm": 5.681210517883301,
      "learning_rate": 0.00017454331172657631,
      "loss": 4.174673461914063,
      "step": 270
    },
    {
      "epoch": 0.32,
      "grad_norm": 5.671685218811035,
      "learning_rate": 0.00017336476134354744,
      "loss": 3.9921218872070314,
      "step": 280
    },
    {
      "epoch": 0.3314285714285714,
      "grad_norm": 6.2818498611450195,
      "learning_rate": 0.00017218621096051857,
      "loss": 4.121313095092773,
      "step": 290
    },
    {
      "epoch": 0.34285714285714286,
      "grad_norm": 5.281393051147461,
      "learning_rate": 0.0001710076605774897,
      "loss": 4.037456512451172,
      "step": 300
    },
    {
      "epoch": 0.34285714285714286,
      "eval_loss": 0.966233491897583,
      "eval_runtime": 189.9725,
      "eval_samples_per_second": 5.264,
      "eval_steps_per_second": 5.264,
      "step": 300
    },
    {
      "epoch": 0.35428571428571426,
      "grad_norm": 5.695072174072266,
      "learning_rate": 0.00016982911019446082,
      "loss": 3.9606529235839845,
      "step": 310
    },
    {
      "epoch": 0.3657142857142857,
      "grad_norm": 5.077876091003418,
      "learning_rate": 0.00016865055981143194,
      "loss": 3.871094512939453,
      "step": 320
    },
    {
      "epoch": 0.37714285714285717,
      "grad_norm": 5.337680339813232,
      "learning_rate": 0.00016747200942840307,
      "loss": 3.84930419921875,
      "step": 330
    },
    {
      "epoch": 0.38857142857142857,
      "grad_norm": 5.603011131286621,
      "learning_rate": 0.0001662934590453742,
      "loss": 3.7293975830078123,
      "step": 340
    },
    {
      "epoch": 0.4,
      "grad_norm": 5.253485679626465,
      "learning_rate": 0.00016511490866234532,
      "loss": 3.993833541870117,
      "step": 350
    },
    {
      "epoch": 0.4114285714285714,
      "grad_norm": 5.056275367736816,
      "learning_rate": 0.00016393635827931645,
      "loss": 3.8682727813720703,
      "step": 360
    },
    {
      "epoch": 0.4228571428571429,
      "grad_norm": 5.209413528442383,
      "learning_rate": 0.00016275780789628758,
      "loss": 3.816153335571289,
      "step": 370
    },
    {
      "epoch": 0.4342857142857143,
      "grad_norm": 5.370687484741211,
      "learning_rate": 0.0001615792575132587,
      "loss": 3.76318359375,
      "step": 380
    },
    {
      "epoch": 0.44571428571428573,
      "grad_norm": 5.17402982711792,
      "learning_rate": 0.00016040070713022983,
      "loss": 3.8684104919433593,
      "step": 390
    },
    {
      "epoch": 0.45714285714285713,
      "grad_norm": 4.992661476135254,
      "learning_rate": 0.00015922215674720095,
      "loss": 3.8093425750732424,
      "step": 400
    },
    {
      "epoch": 0.4685714285714286,
      "grad_norm": 5.006243705749512,
      "learning_rate": 0.00015804360636417208,
      "loss": 3.9194854736328124,
      "step": 410
    },
    {
      "epoch": 0.48,
      "grad_norm": 5.124022006988525,
      "learning_rate": 0.0001568650559811432,
      "loss": 3.857425308227539,
      "step": 420
    },
    {
      "epoch": 0.49142857142857144,
      "grad_norm": 5.376460075378418,
      "learning_rate": 0.00015568650559811433,
      "loss": 3.7050807952880858,
      "step": 430
    },
    {
      "epoch": 0.5028571428571429,
      "grad_norm": 4.906102657318115,
      "learning_rate": 0.00015450795521508546,
      "loss": 3.768130874633789,
      "step": 440
    },
    {
      "epoch": 0.5142857142857142,
      "grad_norm": 5.371730804443359,
      "learning_rate": 0.00015332940483205659,
      "loss": 3.8190208435058595,
      "step": 450
    },
    {
      "epoch": 0.5257142857142857,
      "grad_norm": 5.475566387176514,
      "learning_rate": 0.0001521508544490277,
      "loss": 3.788045120239258,
      "step": 460
    },
    {
      "epoch": 0.5371428571428571,
      "grad_norm": 5.15871524810791,
      "learning_rate": 0.00015097230406599884,
      "loss": 3.6049434661865236,
      "step": 470
    },
    {
      "epoch": 0.5485714285714286,
      "grad_norm": 4.677049160003662,
      "learning_rate": 0.00014979375368296996,
      "loss": 3.5550613403320312,
      "step": 480
    },
    {
      "epoch": 0.56,
      "grad_norm": 4.52909517288208,
      "learning_rate": 0.00014861520329994106,
      "loss": 3.7386096954345702,
      "step": 490
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 4.81787633895874,
      "learning_rate": 0.0001474366529169122,
      "loss": 3.7458377838134767,
      "step": 500
    },
    {
      "epoch": 0.5828571428571429,
      "grad_norm": 5.093353271484375,
      "learning_rate": 0.00014625810253388332,
      "loss": 3.732608413696289,
      "step": 510
    },
    {
      "epoch": 0.5942857142857143,
      "grad_norm": 4.84699010848999,
      "learning_rate": 0.00014507955215085444,
      "loss": 3.5908126831054688,
      "step": 520
    },
    {
      "epoch": 0.6057142857142858,
      "grad_norm": 5.440133571624756,
      "learning_rate": 0.00014390100176782557,
      "loss": 3.6844512939453127,
      "step": 530
    },
    {
      "epoch": 0.6171428571428571,
      "grad_norm": 4.786667346954346,
      "learning_rate": 0.0001427224513847967,
      "loss": 3.55334587097168,
      "step": 540
    },
    {
      "epoch": 0.6285714285714286,
      "grad_norm": 4.713446140289307,
      "learning_rate": 0.00014154390100176785,
      "loss": 3.671040344238281,
      "step": 550
    },
    {
      "epoch": 0.64,
      "grad_norm": 4.542754173278809,
      "learning_rate": 0.00014036535061873897,
      "loss": 3.565074920654297,
      "step": 560
    },
    {
      "epoch": 0.6514285714285715,
      "grad_norm": 4.633498668670654,
      "learning_rate": 0.0001391868002357101,
      "loss": 3.5600765228271483,
      "step": 570
    },
    {
      "epoch": 0.6628571428571428,
      "grad_norm": 5.180357456207275,
      "learning_rate": 0.00013800824985268123,
      "loss": 3.590932846069336,
      "step": 580
    },
    {
      "epoch": 0.6742857142857143,
      "grad_norm": 5.013422966003418,
      "learning_rate": 0.00013682969946965235,
      "loss": 3.543553924560547,
      "step": 590
    },
    {
      "epoch": 0.6857142857142857,
      "grad_norm": 4.720696449279785,
      "learning_rate": 0.00013565114908662345,
      "loss": 3.6398658752441406,
      "step": 600
    },
    {
      "epoch": 0.6857142857142857,
      "eval_loss": 0.8644659519195557,
      "eval_runtime": 188.2427,
      "eval_samples_per_second": 5.312,
      "eval_steps_per_second": 5.312,
      "step": 600
    },
    {
      "epoch": 0.6971428571428572,
      "grad_norm": 4.762374401092529,
      "learning_rate": 0.00013447259870359458,
      "loss": 3.5501239776611326,
      "step": 610
    },
    {
      "epoch": 0.7085714285714285,
      "grad_norm": 4.8993916511535645,
      "learning_rate": 0.0001332940483205657,
      "loss": 3.4655189514160156,
      "step": 620
    },
    {
      "epoch": 0.72,
      "grad_norm": 4.667840003967285,
      "learning_rate": 0.00013211549793753683,
      "loss": 3.559103775024414,
      "step": 630
    },
    {
      "epoch": 0.7314285714285714,
      "grad_norm": 4.983627796173096,
      "learning_rate": 0.00013093694755450796,
      "loss": 3.538484573364258,
      "step": 640
    },
    {
      "epoch": 0.7428571428571429,
      "grad_norm": 4.6870222091674805,
      "learning_rate": 0.00012975839717147908,
      "loss": 3.434688949584961,
      "step": 650
    },
    {
      "epoch": 0.7542857142857143,
      "grad_norm": 4.765181064605713,
      "learning_rate": 0.0001285798467884502,
      "loss": 3.3901046752929687,
      "step": 660
    },
    {
      "epoch": 0.7657142857142857,
      "grad_norm": 4.89863920211792,
      "learning_rate": 0.00012740129640542133,
      "loss": 3.4148059844970704,
      "step": 670
    },
    {
      "epoch": 0.7771428571428571,
      "grad_norm": 5.234548091888428,
      "learning_rate": 0.00012622274602239246,
      "loss": 3.603266143798828,
      "step": 680
    },
    {
      "epoch": 0.7885714285714286,
      "grad_norm": 4.550261974334717,
      "learning_rate": 0.0001250441956393636,
      "loss": 3.397371292114258,
      "step": 690
    },
    {
      "epoch": 0.8,
      "grad_norm": 5.03040075302124,
      "learning_rate": 0.0001238656452563347,
      "loss": 3.326694869995117,
      "step": 700
    },
    {
      "epoch": 0.8114285714285714,
      "grad_norm": 4.639382839202881,
      "learning_rate": 0.00012268709487330584,
      "loss": 3.4932540893554687,
      "step": 710
    },
    {
      "epoch": 0.8228571428571428,
      "grad_norm": 4.673860549926758,
      "learning_rate": 0.00012150854449027695,
      "loss": 3.356616973876953,
      "step": 720
    },
    {
      "epoch": 0.8342857142857143,
      "grad_norm": 4.805152893066406,
      "learning_rate": 0.0001203299941072481,
      "loss": 3.461750793457031,
      "step": 730
    },
    {
      "epoch": 0.8457142857142858,
      "grad_norm": 4.468682289123535,
      "learning_rate": 0.00011915144372421923,
      "loss": 3.47204704284668,
      "step": 740
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 4.596803665161133,
      "learning_rate": 0.00011797289334119034,
      "loss": 3.44429931640625,
      "step": 750
    },
    {
      "epoch": 0.8685714285714285,
      "grad_norm": 4.497738838195801,
      "learning_rate": 0.00011679434295816147,
      "loss": 3.41302490234375,
      "step": 760
    },
    {
      "epoch": 0.88,
      "grad_norm": 4.756861209869385,
      "learning_rate": 0.0001156157925751326,
      "loss": 3.3843093872070313,
      "step": 770
    },
    {
      "epoch": 0.8914285714285715,
      "grad_norm": 4.649412631988525,
      "learning_rate": 0.00011443724219210372,
      "loss": 3.2944171905517576,
      "step": 780
    },
    {
      "epoch": 0.9028571428571428,
      "grad_norm": 4.758933067321777,
      "learning_rate": 0.00011325869180907485,
      "loss": 3.371187210083008,
      "step": 790
    },
    {
      "epoch": 0.9142857142857143,
      "grad_norm": 4.970985412597656,
      "learning_rate": 0.00011208014142604597,
      "loss": 3.2973373413085936,
      "step": 800
    },
    {
      "epoch": 0.9257142857142857,
      "grad_norm": 4.600860595703125,
      "learning_rate": 0.0001109015910430171,
      "loss": 3.363826370239258,
      "step": 810
    },
    {
      "epoch": 0.9371428571428572,
      "grad_norm": 4.5820722579956055,
      "learning_rate": 0.00010972304065998821,
      "loss": 3.3119132995605467,
      "step": 820
    },
    {
      "epoch": 0.9485714285714286,
      "grad_norm": 4.491682529449463,
      "learning_rate": 0.00010854449027695934,
      "loss": 3.3644573211669924,
      "step": 830
    },
    {
      "epoch": 0.96,
      "grad_norm": 4.860896110534668,
      "learning_rate": 0.00010736593989393047,
      "loss": 3.374749755859375,
      "step": 840
    },
    {
      "epoch": 0.9714285714285714,
      "grad_norm": 4.7770891189575195,
      "learning_rate": 0.00010618738951090159,
      "loss": 3.2817356109619142,
      "step": 850
    },
    {
      "epoch": 0.9828571428571429,
      "grad_norm": 4.682967662811279,
      "learning_rate": 0.00010500883912787272,
      "loss": 3.367177200317383,
      "step": 860
    },
    {
      "epoch": 0.9942857142857143,
      "grad_norm": 4.5394158363342285,
      "learning_rate": 0.00010383028874484384,
      "loss": 3.3502220153808593,
      "step": 870
    },
    {
      "epoch": 1.0057142857142858,
      "grad_norm": 4.440478324890137,
      "learning_rate": 0.00010265173836181497,
      "loss": 2.993194007873535,
      "step": 880
    },
    {
      "epoch": 1.0171428571428571,
      "grad_norm": 4.419264316558838,
      "learning_rate": 0.00010147318797878608,
      "loss": 2.858263397216797,
      "step": 890
    },
    {
      "epoch": 1.0285714285714285,
      "grad_norm": 4.7429327964782715,
      "learning_rate": 0.00010029463759575721,
      "loss": 2.893912696838379,
      "step": 900
    },
    {
      "epoch": 1.0285714285714285,
      "eval_loss": 0.8113172650337219,
      "eval_runtime": 187.5319,
      "eval_samples_per_second": 5.332,
      "eval_steps_per_second": 5.332,
      "step": 900
    }
  ],
  "logging_steps": 10,
  "max_steps": 1750,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 300,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7.385003465695887e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}