{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9941520467836256,
  "eval_steps": 500,
  "global_step": 576,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05198180636777128,
      "grad_norm": 3.0779116014443817,
      "learning_rate": 5e-06,
      "loss": 0.7363,
      "step": 10
    },
    {
      "epoch": 0.10396361273554255,
      "grad_norm": 0.8378025538905853,
      "learning_rate": 5e-06,
      "loss": 0.6574,
      "step": 20
    },
    {
      "epoch": 0.15594541910331383,
      "grad_norm": 1.0326593461431761,
      "learning_rate": 5e-06,
      "loss": 0.6354,
      "step": 30
    },
    {
      "epoch": 0.2079272254710851,
      "grad_norm": 0.8798298291096642,
      "learning_rate": 5e-06,
      "loss": 0.6258,
      "step": 40
    },
    {
      "epoch": 0.2599090318388564,
      "grad_norm": 0.7652583221522038,
      "learning_rate": 5e-06,
      "loss": 0.6081,
      "step": 50
    },
    {
      "epoch": 0.31189083820662766,
      "grad_norm": 0.8092758063286689,
      "learning_rate": 5e-06,
      "loss": 0.5957,
      "step": 60
    },
    {
      "epoch": 0.36387264457439894,
      "grad_norm": 0.6377593975168986,
      "learning_rate": 5e-06,
      "loss": 0.5951,
      "step": 70
    },
    {
      "epoch": 0.4158544509421702,
      "grad_norm": 0.8604320761158406,
      "learning_rate": 5e-06,
      "loss": 0.5853,
      "step": 80
    },
    {
      "epoch": 0.4678362573099415,
      "grad_norm": 0.5659478868871771,
      "learning_rate": 5e-06,
      "loss": 0.5736,
      "step": 90
    },
    {
      "epoch": 0.5198180636777128,
      "grad_norm": 0.6666095975840195,
      "learning_rate": 5e-06,
      "loss": 0.5681,
      "step": 100
    },
    {
      "epoch": 0.571799870045484,
      "grad_norm": 0.7108362162785513,
      "learning_rate": 5e-06,
      "loss": 0.5752,
      "step": 110
    },
    {
      "epoch": 0.6237816764132553,
      "grad_norm": 0.5055975881819947,
      "learning_rate": 5e-06,
      "loss": 0.5661,
      "step": 120
    },
    {
      "epoch": 0.6757634827810266,
      "grad_norm": 0.5122594507537827,
      "learning_rate": 5e-06,
      "loss": 0.5724,
      "step": 130
    },
    {
      "epoch": 0.7277452891487979,
      "grad_norm": 0.6645034900787495,
      "learning_rate": 5e-06,
      "loss": 0.5642,
      "step": 140
    },
    {
      "epoch": 0.7797270955165692,
      "grad_norm": 0.4688383993640717,
      "learning_rate": 5e-06,
      "loss": 0.563,
      "step": 150
    },
    {
      "epoch": 0.8317089018843404,
      "grad_norm": 0.5311685960170432,
      "learning_rate": 5e-06,
      "loss": 0.5636,
      "step": 160
    },
    {
      "epoch": 0.8836907082521117,
      "grad_norm": 0.5275216737667017,
      "learning_rate": 5e-06,
      "loss": 0.5622,
      "step": 170
    },
    {
      "epoch": 0.935672514619883,
      "grad_norm": 0.49741573059254773,
      "learning_rate": 5e-06,
      "loss": 0.5604,
      "step": 180
    },
    {
      "epoch": 0.9876543209876543,
      "grad_norm": 0.6752670080708038,
      "learning_rate": 5e-06,
      "loss": 0.5606,
      "step": 190
    },
    {
      "epoch": 0.9980506822612085,
      "eval_loss": 0.5604754090309143,
      "eval_runtime": 134.8008,
      "eval_samples_per_second": 38.457,
      "eval_steps_per_second": 0.601,
      "step": 192
    },
    {
      "epoch": 1.0396361273554255,
      "grad_norm": 0.4942854982934459,
      "learning_rate": 5e-06,
      "loss": 0.5581,
      "step": 200
    },
    {
      "epoch": 1.0916179337231968,
      "grad_norm": 0.47133396673557604,
      "learning_rate": 5e-06,
      "loss": 0.5131,
      "step": 210
    },
    {
      "epoch": 1.143599740090968,
      "grad_norm": 0.6706210343356649,
      "learning_rate": 5e-06,
      "loss": 0.5108,
      "step": 220
    },
    {
      "epoch": 1.1955815464587394,
      "grad_norm": 0.4990475384686441,
      "learning_rate": 5e-06,
      "loss": 0.5127,
      "step": 230
    },
    {
      "epoch": 1.2475633528265107,
      "grad_norm": 0.6948432011959483,
      "learning_rate": 5e-06,
      "loss": 0.516,
      "step": 240
    },
    {
      "epoch": 1.299545159194282,
      "grad_norm": 0.4617478562618637,
      "learning_rate": 5e-06,
      "loss": 0.5124,
      "step": 250
    },
    {
      "epoch": 1.3515269655620532,
      "grad_norm": 0.40496266862562025,
      "learning_rate": 5e-06,
      "loss": 0.5169,
      "step": 260
    },
    {
      "epoch": 1.4035087719298245,
      "grad_norm": 0.4441934839800834,
      "learning_rate": 5e-06,
      "loss": 0.5153,
      "step": 270
    },
    {
      "epoch": 1.4554905782975958,
      "grad_norm": 0.5551575728453572,
      "learning_rate": 5e-06,
      "loss": 0.5153,
      "step": 280
    },
    {
      "epoch": 1.507472384665367,
      "grad_norm": 0.6109195782274622,
      "learning_rate": 5e-06,
      "loss": 0.5095,
      "step": 290
    },
    {
      "epoch": 1.5594541910331383,
      "grad_norm": 0.501987852874291,
      "learning_rate": 5e-06,
      "loss": 0.5134,
      "step": 300
    },
    {
      "epoch": 1.6114359974009096,
      "grad_norm": 0.42511170270357557,
      "learning_rate": 5e-06,
      "loss": 0.5165,
      "step": 310
    },
    {
      "epoch": 1.6634178037686809,
      "grad_norm": 0.49289651016151736,
      "learning_rate": 5e-06,
      "loss": 0.5096,
      "step": 320
    },
    {
      "epoch": 1.7153996101364521,
      "grad_norm": 0.4308251737050951,
      "learning_rate": 5e-06,
      "loss": 0.5138,
      "step": 330
    },
    {
      "epoch": 1.7673814165042234,
      "grad_norm": 0.5157909639859894,
      "learning_rate": 5e-06,
      "loss": 0.5171,
      "step": 340
    },
    {
      "epoch": 1.8193632228719947,
      "grad_norm": 0.5188996520268809,
      "learning_rate": 5e-06,
      "loss": 0.5038,
      "step": 350
    },
    {
      "epoch": 1.871345029239766,
      "grad_norm": 0.4609663557546815,
      "learning_rate": 5e-06,
      "loss": 0.5135,
      "step": 360
    },
    {
      "epoch": 1.9233268356075373,
      "grad_norm": 0.5130233449439461,
      "learning_rate": 5e-06,
      "loss": 0.4969,
      "step": 370
    },
    {
      "epoch": 1.9753086419753085,
      "grad_norm": 0.47208230393580913,
      "learning_rate": 5e-06,
      "loss": 0.5116,
      "step": 380
    },
    {
      "epoch": 1.996101364522417,
      "eval_loss": 0.5507273077964783,
      "eval_runtime": 131.5009,
      "eval_samples_per_second": 39.422,
      "eval_steps_per_second": 0.616,
      "step": 384
    },
    {
      "epoch": 2.02729044834308,
      "grad_norm": 0.6127372201164472,
      "learning_rate": 5e-06,
      "loss": 0.5155,
      "step": 390
    },
    {
      "epoch": 2.079272254710851,
      "grad_norm": 0.7411594983025063,
      "learning_rate": 5e-06,
      "loss": 0.4578,
      "step": 400
    },
    {
      "epoch": 2.1312540610786224,
      "grad_norm": 0.7428780446595316,
      "learning_rate": 5e-06,
      "loss": 0.4704,
      "step": 410
    },
    {
      "epoch": 2.1832358674463936,
      "grad_norm": 0.5200617783769791,
      "learning_rate": 5e-06,
      "loss": 0.4637,
      "step": 420
    },
    {
      "epoch": 2.235217673814165,
      "grad_norm": 0.6498653019719489,
      "learning_rate": 5e-06,
      "loss": 0.4622,
      "step": 430
    },
    {
      "epoch": 2.287199480181936,
      "grad_norm": 0.46276007763796656,
      "learning_rate": 5e-06,
      "loss": 0.4607,
      "step": 440
    },
    {
      "epoch": 2.3391812865497075,
      "grad_norm": 0.5996673377937245,
      "learning_rate": 5e-06,
      "loss": 0.4622,
      "step": 450
    },
    {
      "epoch": 2.3911630929174787,
      "grad_norm": 0.49488878939584074,
      "learning_rate": 5e-06,
      "loss": 0.4682,
      "step": 460
    },
    {
      "epoch": 2.44314489928525,
      "grad_norm": 0.4643784900227937,
      "learning_rate": 5e-06,
      "loss": 0.4594,
      "step": 470
    },
    {
      "epoch": 2.4951267056530213,
      "grad_norm": 0.44786233784344404,
      "learning_rate": 5e-06,
      "loss": 0.4675,
      "step": 480
    },
    {
      "epoch": 2.5471085120207926,
      "grad_norm": 0.6126679422066866,
      "learning_rate": 5e-06,
      "loss": 0.4659,
      "step": 490
    },
    {
      "epoch": 2.599090318388564,
      "grad_norm": 0.530097732347591,
      "learning_rate": 5e-06,
      "loss": 0.4588,
      "step": 500
    },
    {
      "epoch": 2.651072124756335,
      "grad_norm": 0.5516810761121376,
      "learning_rate": 5e-06,
      "loss": 0.4634,
      "step": 510
    },
    {
      "epoch": 2.7030539311241064,
      "grad_norm": 0.5437842249870648,
      "learning_rate": 5e-06,
      "loss": 0.4642,
      "step": 520
    },
    {
      "epoch": 2.7550357374918777,
      "grad_norm": 0.5248669258621385,
      "learning_rate": 5e-06,
      "loss": 0.4658,
      "step": 530
    },
    {
      "epoch": 2.807017543859649,
      "grad_norm": 0.4756943098859436,
      "learning_rate": 5e-06,
      "loss": 0.4712,
      "step": 540
    },
    {
      "epoch": 2.8589993502274202,
      "grad_norm": 0.5165218818933938,
      "learning_rate": 5e-06,
      "loss": 0.4672,
      "step": 550
    },
    {
      "epoch": 2.9109811565951915,
      "grad_norm": 0.5304020568782208,
      "learning_rate": 5e-06,
      "loss": 0.4682,
      "step": 560
    },
    {
      "epoch": 2.962962962962963,
      "grad_norm": 0.5954622072015786,
      "learning_rate": 5e-06,
      "loss": 0.4569,
      "step": 570
    },
    {
      "epoch": 2.9941520467836256,
      "eval_loss": 0.5534155368804932,
      "eval_runtime": 134.2301,
      "eval_samples_per_second": 38.62,
      "eval_steps_per_second": 0.603,
      "step": 576
    },
    {
      "epoch": 2.9941520467836256,
      "step": 576,
      "total_flos": 964612073717760.0,
      "train_loss": 0.5243237420088716,
      "train_runtime": 19210.653,
      "train_samples_per_second": 15.379,
      "train_steps_per_second": 0.03
    }
  ],
  "logging_steps": 10,
  "max_steps": 576,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 964612073717760.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}