{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 975,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03076923076923077,
      "grad_norm": 2.3324186560475026,
      "learning_rate": 5e-06,
      "loss": 0.9796,
      "step": 10
    },
    {
      "epoch": 0.06153846153846154,
      "grad_norm": 1.0624345768919523,
      "learning_rate": 5e-06,
      "loss": 0.8629,
      "step": 20
    },
    {
      "epoch": 0.09230769230769231,
      "grad_norm": 1.0856708355070386,
      "learning_rate": 5e-06,
      "loss": 0.8232,
      "step": 30
    },
    {
      "epoch": 0.12307692307692308,
      "grad_norm": 1.3950701294784864,
      "learning_rate": 5e-06,
      "loss": 0.8034,
      "step": 40
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 0.9965613643303693,
      "learning_rate": 5e-06,
      "loss": 0.7826,
      "step": 50
    },
    {
      "epoch": 0.18461538461538463,
      "grad_norm": 0.835067800401052,
      "learning_rate": 5e-06,
      "loss": 0.7723,
      "step": 60
    },
    {
      "epoch": 0.2153846153846154,
      "grad_norm": 0.781647957469068,
      "learning_rate": 5e-06,
      "loss": 0.767,
      "step": 70
    },
    {
      "epoch": 0.24615384615384617,
      "grad_norm": 1.1712549789017386,
      "learning_rate": 5e-06,
      "loss": 0.7559,
      "step": 80
    },
    {
      "epoch": 0.27692307692307694,
      "grad_norm": 1.0092940618549995,
      "learning_rate": 5e-06,
      "loss": 0.7469,
      "step": 90
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 0.5624307958541879,
      "learning_rate": 5e-06,
      "loss": 0.7414,
      "step": 100
    },
    {
      "epoch": 0.3384615384615385,
      "grad_norm": 0.5739018842323079,
      "learning_rate": 5e-06,
      "loss": 0.7402,
      "step": 110
    },
    {
      "epoch": 0.36923076923076925,
      "grad_norm": 0.5881069634317425,
      "learning_rate": 5e-06,
      "loss": 0.7409,
      "step": 120
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.7381656504744467,
      "learning_rate": 5e-06,
      "loss": 0.7376,
      "step": 130
    },
    {
      "epoch": 0.4307692307692308,
      "grad_norm": 0.5677544829127201,
      "learning_rate": 5e-06,
      "loss": 0.7298,
      "step": 140
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 0.7988060295240222,
      "learning_rate": 5e-06,
      "loss": 0.734,
      "step": 150
    },
    {
      "epoch": 0.49230769230769234,
      "grad_norm": 0.6718712556340437,
      "learning_rate": 5e-06,
      "loss": 0.7301,
      "step": 160
    },
    {
      "epoch": 0.5230769230769231,
      "grad_norm": 0.6738769041614976,
      "learning_rate": 5e-06,
      "loss": 0.7316,
      "step": 170
    },
    {
      "epoch": 0.5538461538461539,
      "grad_norm": 0.5838925373137313,
      "learning_rate": 5e-06,
      "loss": 0.7253,
      "step": 180
    },
    {
      "epoch": 0.5846153846153846,
      "grad_norm": 0.6725428665358475,
      "learning_rate": 5e-06,
      "loss": 0.7241,
      "step": 190
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.5603996763654583,
      "learning_rate": 5e-06,
      "loss": 0.7228,
      "step": 200
    },
    {
      "epoch": 0.6461538461538462,
      "grad_norm": 0.7449029762390439,
      "learning_rate": 5e-06,
      "loss": 0.7204,
      "step": 210
    },
    {
      "epoch": 0.676923076923077,
      "grad_norm": 0.5236476916978245,
      "learning_rate": 5e-06,
      "loss": 0.717,
      "step": 220
    },
    {
      "epoch": 0.7076923076923077,
      "grad_norm": 0.6865974040573393,
      "learning_rate": 5e-06,
      "loss": 0.7134,
      "step": 230
    },
    {
      "epoch": 0.7384615384615385,
      "grad_norm": 0.6717184095724111,
      "learning_rate": 5e-06,
      "loss": 0.7159,
      "step": 240
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.8235556500295945,
      "learning_rate": 5e-06,
      "loss": 0.711,
      "step": 250
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5131501040264781,
      "learning_rate": 5e-06,
      "loss": 0.7165,
      "step": 260
    },
    {
      "epoch": 0.8307692307692308,
      "grad_norm": 0.5193918430229191,
      "learning_rate": 5e-06,
      "loss": 0.7055,
      "step": 270
    },
    {
      "epoch": 0.8615384615384616,
      "grad_norm": 0.5456897002979056,
      "learning_rate": 5e-06,
      "loss": 0.711,
      "step": 280
    },
    {
      "epoch": 0.8923076923076924,
      "grad_norm": 0.5750072559309753,
      "learning_rate": 5e-06,
      "loss": 0.7084,
      "step": 290
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 0.6116232515773206,
      "learning_rate": 5e-06,
      "loss": 0.7128,
      "step": 300
    },
    {
      "epoch": 0.9538461538461539,
      "grad_norm": 0.5681793375147939,
      "learning_rate": 5e-06,
      "loss": 0.7086,
      "step": 310
    },
    {
      "epoch": 0.9846153846153847,
      "grad_norm": 0.6343675909171526,
      "learning_rate": 5e-06,
      "loss": 0.7104,
      "step": 320
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.7103046774864197,
      "eval_runtime": 341.5784,
      "eval_samples_per_second": 25.637,
      "eval_steps_per_second": 0.401,
      "step": 325
    },
    {
      "epoch": 1.0153846153846153,
      "grad_norm": 0.7692870192954333,
      "learning_rate": 5e-06,
      "loss": 0.6897,
      "step": 330
    },
    {
      "epoch": 1.0461538461538462,
      "grad_norm": 0.5505626206403108,
      "learning_rate": 5e-06,
      "loss": 0.6636,
      "step": 340
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 0.5824910302345806,
      "learning_rate": 5e-06,
      "loss": 0.665,
      "step": 350
    },
    {
      "epoch": 1.1076923076923078,
      "grad_norm": 0.6417795632315966,
      "learning_rate": 5e-06,
      "loss": 0.664,
      "step": 360
    },
    {
      "epoch": 1.1384615384615384,
      "grad_norm": 0.6909469669994084,
      "learning_rate": 5e-06,
      "loss": 0.6634,
      "step": 370
    },
    {
      "epoch": 1.1692307692307693,
      "grad_norm": 0.5616732809499269,
      "learning_rate": 5e-06,
      "loss": 0.6653,
      "step": 380
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.567347803253029,
      "learning_rate": 5e-06,
      "loss": 0.6605,
      "step": 390
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.6020788318645328,
      "learning_rate": 5e-06,
      "loss": 0.6649,
      "step": 400
    },
    {
      "epoch": 1.2615384615384615,
      "grad_norm": 0.6247830402384781,
      "learning_rate": 5e-06,
      "loss": 0.6662,
      "step": 410
    },
    {
      "epoch": 1.2923076923076924,
      "grad_norm": 0.7223035560412274,
      "learning_rate": 5e-06,
      "loss": 0.6631,
      "step": 420
    },
    {
      "epoch": 1.323076923076923,
      "grad_norm": 0.7277266215407076,
      "learning_rate": 5e-06,
      "loss": 0.6596,
      "step": 430
    },
    {
      "epoch": 1.353846153846154,
      "grad_norm": 0.5531058222331595,
      "learning_rate": 5e-06,
      "loss": 0.6659,
      "step": 440
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.7560378640801586,
      "learning_rate": 5e-06,
      "loss": 0.6654,
      "step": 450
    },
    {
      "epoch": 1.4153846153846155,
      "grad_norm": 0.6002650702602214,
      "learning_rate": 5e-06,
      "loss": 0.6602,
      "step": 460
    },
    {
      "epoch": 1.4461538461538461,
      "grad_norm": 0.5446217104790005,
      "learning_rate": 5e-06,
      "loss": 0.6633,
      "step": 470
    },
    {
      "epoch": 1.476923076923077,
      "grad_norm": 0.6427854645725265,
      "learning_rate": 5e-06,
      "loss": 0.6606,
      "step": 480
    },
    {
      "epoch": 1.5076923076923077,
      "grad_norm": 0.5777163615379136,
      "learning_rate": 5e-06,
      "loss": 0.6619,
      "step": 490
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.5645927733372786,
      "learning_rate": 5e-06,
      "loss": 0.6613,
      "step": 500
    },
    {
      "epoch": 1.5692307692307692,
      "grad_norm": 0.5877237236567665,
      "learning_rate": 5e-06,
      "loss": 0.6666,
      "step": 510
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.7318508117188922,
      "learning_rate": 5e-06,
      "loss": 0.6657,
      "step": 520
    },
    {
      "epoch": 1.6307692307692307,
      "grad_norm": 0.5337461204074161,
      "learning_rate": 5e-06,
      "loss": 0.6678,
      "step": 530
    },
    {
      "epoch": 1.6615384615384614,
      "grad_norm": 0.6743103368990547,
      "learning_rate": 5e-06,
      "loss": 0.6593,
      "step": 540
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 0.7414434681428989,
      "learning_rate": 5e-06,
      "loss": 0.6575,
      "step": 550
    },
    {
      "epoch": 1.7230769230769232,
      "grad_norm": 0.8444136346063817,
      "learning_rate": 5e-06,
      "loss": 0.6631,
      "step": 560
    },
    {
      "epoch": 1.7538461538461538,
      "grad_norm": 0.5509955001744837,
      "learning_rate": 5e-06,
      "loss": 0.6592,
      "step": 570
    },
    {
      "epoch": 1.7846153846153845,
      "grad_norm": 0.6630480503790037,
      "learning_rate": 5e-06,
      "loss": 0.6594,
      "step": 580
    },
    {
      "epoch": 1.8153846153846154,
      "grad_norm": 0.5120012670268558,
      "learning_rate": 5e-06,
      "loss": 0.6568,
      "step": 590
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.5752130113945352,
      "learning_rate": 5e-06,
      "loss": 0.6599,
      "step": 600
    },
    {
      "epoch": 1.876923076923077,
      "grad_norm": 0.5835189539855182,
      "learning_rate": 5e-06,
      "loss": 0.6591,
      "step": 610
    },
    {
      "epoch": 1.9076923076923076,
      "grad_norm": 0.5603728195599548,
      "learning_rate": 5e-06,
      "loss": 0.6629,
      "step": 620
    },
    {
      "epoch": 1.9384615384615385,
      "grad_norm": 0.6856529088244573,
      "learning_rate": 5e-06,
      "loss": 0.6586,
      "step": 630
    },
    {
      "epoch": 1.9692307692307693,
      "grad_norm": 0.5160023610012462,
      "learning_rate": 5e-06,
      "loss": 0.6603,
      "step": 640
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6268329172925445,
      "learning_rate": 5e-06,
      "loss": 0.658,
      "step": 650
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6978012323379517,
      "eval_runtime": 343.0061,
      "eval_samples_per_second": 25.53,
      "eval_steps_per_second": 0.399,
      "step": 650
    },
    {
      "epoch": 2.0307692307692307,
      "grad_norm": 0.7968437785991972,
      "learning_rate": 5e-06,
      "loss": 0.6088,
      "step": 660
    },
    {
      "epoch": 2.0615384615384613,
      "grad_norm": 1.1539294867806145,
      "learning_rate": 5e-06,
      "loss": 0.6083,
      "step": 670
    },
    {
      "epoch": 2.0923076923076924,
      "grad_norm": 0.9623521768346174,
      "learning_rate": 5e-06,
      "loss": 0.6149,
      "step": 680
    },
    {
      "epoch": 2.123076923076923,
      "grad_norm": 0.6040507943945268,
      "learning_rate": 5e-06,
      "loss": 0.6164,
      "step": 690
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.5851170034060879,
      "learning_rate": 5e-06,
      "loss": 0.6115,
      "step": 700
    },
    {
      "epoch": 2.184615384615385,
      "grad_norm": 0.600096702685487,
      "learning_rate": 5e-06,
      "loss": 0.6153,
      "step": 710
    },
    {
      "epoch": 2.2153846153846155,
      "grad_norm": 0.5589523777176305,
      "learning_rate": 5e-06,
      "loss": 0.6133,
      "step": 720
    },
    {
      "epoch": 2.246153846153846,
      "grad_norm": 0.5907169270955183,
      "learning_rate": 5e-06,
      "loss": 0.6131,
      "step": 730
    },
    {
      "epoch": 2.276923076923077,
      "grad_norm": 0.6628827223280479,
      "learning_rate": 5e-06,
      "loss": 0.6159,
      "step": 740
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.6215774766516785,
      "learning_rate": 5e-06,
      "loss": 0.6149,
      "step": 750
    },
    {
      "epoch": 2.3384615384615386,
      "grad_norm": 0.6649815302458084,
      "learning_rate": 5e-06,
      "loss": 0.6132,
      "step": 760
    },
    {
      "epoch": 2.3692307692307693,
      "grad_norm": 0.699779615122456,
      "learning_rate": 5e-06,
      "loss": 0.6128,
      "step": 770
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.546711048417698,
      "learning_rate": 5e-06,
      "loss": 0.6164,
      "step": 780
    },
    {
      "epoch": 2.430769230769231,
      "grad_norm": 0.6327118907910535,
      "learning_rate": 5e-06,
      "loss": 0.623,
      "step": 790
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.5174866030508046,
      "learning_rate": 5e-06,
      "loss": 0.6166,
      "step": 800
    },
    {
      "epoch": 2.4923076923076923,
      "grad_norm": 0.5497949972140503,
      "learning_rate": 5e-06,
      "loss": 0.615,
      "step": 810
    },
    {
      "epoch": 2.523076923076923,
      "grad_norm": 0.5993580492245987,
      "learning_rate": 5e-06,
      "loss": 0.614,
      "step": 820
    },
    {
      "epoch": 2.5538461538461537,
      "grad_norm": 0.6442768713339674,
      "learning_rate": 5e-06,
      "loss": 0.6222,
      "step": 830
    },
    {
      "epoch": 2.5846153846153848,
      "grad_norm": 0.6428843121497717,
      "learning_rate": 5e-06,
      "loss": 0.617,
      "step": 840
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.6117392354746288,
      "learning_rate": 5e-06,
      "loss": 0.62,
      "step": 850
    },
    {
      "epoch": 2.646153846153846,
      "grad_norm": 0.7180295224765794,
      "learning_rate": 5e-06,
      "loss": 0.619,
      "step": 860
    },
    {
      "epoch": 2.676923076923077,
      "grad_norm": 0.6501140616713876,
      "learning_rate": 5e-06,
      "loss": 0.6164,
      "step": 870
    },
    {
      "epoch": 2.707692307692308,
      "grad_norm": 0.5412288061836285,
      "learning_rate": 5e-06,
      "loss": 0.6218,
      "step": 880
    },
    {
      "epoch": 2.7384615384615385,
      "grad_norm": 0.7302614499001912,
      "learning_rate": 5e-06,
      "loss": 0.6176,
      "step": 890
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.6661401784715161,
      "learning_rate": 5e-06,
      "loss": 0.6217,
      "step": 900
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.7263597509162613,
      "learning_rate": 5e-06,
      "loss": 0.6154,
      "step": 910
    },
    {
      "epoch": 2.830769230769231,
      "grad_norm": 0.7506085991128838,
      "learning_rate": 5e-06,
      "loss": 0.6189,
      "step": 920
    },
    {
      "epoch": 2.8615384615384616,
      "grad_norm": 0.5917421117589724,
      "learning_rate": 5e-06,
      "loss": 0.6211,
      "step": 930
    },
    {
      "epoch": 2.8923076923076922,
      "grad_norm": 0.5306708247183923,
      "learning_rate": 5e-06,
      "loss": 0.6175,
      "step": 940
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.6648394376559398,
      "learning_rate": 5e-06,
      "loss": 0.6222,
      "step": 950
    },
    {
      "epoch": 2.953846153846154,
      "grad_norm": 0.5459019838181679,
      "learning_rate": 5e-06,
      "loss": 0.6174,
      "step": 960
    },
    {
      "epoch": 2.9846153846153847,
      "grad_norm": 0.6951553333219783,
      "learning_rate": 5e-06,
      "loss": 0.6176,
      "step": 970
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6986876726150513,
      "eval_runtime": 342.0684,
      "eval_samples_per_second": 25.6,
      "eval_steps_per_second": 0.401,
      "step": 975
    },
    {
      "epoch": 3.0,
      "step": 975,
      "total_flos": 1632951934648320.0,
      "train_loss": 0.6751218311603253,
      "train_runtime": 57099.1896,
      "train_samples_per_second": 8.741,
      "train_steps_per_second": 0.017
    }
  ],
  "logging_steps": 10,
  "max_steps": 975,
  "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": 1632951934648320.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}