{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 822,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0364963503649635,
      "grad_norm": 3.2654903919327523,
      "learning_rate": 5e-06,
      "loss": 1.0419,
      "step": 10
    },
    {
      "epoch": 0.072992700729927,
      "grad_norm": 1.5482985205580984,
      "learning_rate": 5e-06,
      "loss": 0.9117,
      "step": 20
    },
    {
      "epoch": 0.10948905109489052,
      "grad_norm": 1.2069486973191534,
      "learning_rate": 5e-06,
      "loss": 0.8752,
      "step": 30
    },
    {
      "epoch": 0.145985401459854,
      "grad_norm": 1.2169049310832125,
      "learning_rate": 5e-06,
      "loss": 0.8479,
      "step": 40
    },
    {
      "epoch": 0.18248175182481752,
      "grad_norm": 1.0461259998073966,
      "learning_rate": 5e-06,
      "loss": 0.8276,
      "step": 50
    },
    {
      "epoch": 0.21897810218978103,
      "grad_norm": 0.9136364083656688,
      "learning_rate": 5e-06,
      "loss": 0.8144,
      "step": 60
    },
    {
      "epoch": 0.25547445255474455,
      "grad_norm": 1.0016252258412965,
      "learning_rate": 5e-06,
      "loss": 0.8029,
      "step": 70
    },
    {
      "epoch": 0.291970802919708,
      "grad_norm": 0.7256070366121048,
      "learning_rate": 5e-06,
      "loss": 0.7984,
      "step": 80
    },
    {
      "epoch": 0.3284671532846715,
      "grad_norm": 0.8059560568156144,
      "learning_rate": 5e-06,
      "loss": 0.7845,
      "step": 90
    },
    {
      "epoch": 0.36496350364963503,
      "grad_norm": 1.0375230892576275,
      "learning_rate": 5e-06,
      "loss": 0.784,
      "step": 100
    },
    {
      "epoch": 0.40145985401459855,
      "grad_norm": 0.6429193748930873,
      "learning_rate": 5e-06,
      "loss": 0.7786,
      "step": 110
    },
    {
      "epoch": 0.43795620437956206,
      "grad_norm": 0.6387684294515616,
      "learning_rate": 5e-06,
      "loss": 0.7741,
      "step": 120
    },
    {
      "epoch": 0.4744525547445255,
      "grad_norm": 0.7647783012503196,
      "learning_rate": 5e-06,
      "loss": 0.7706,
      "step": 130
    },
    {
      "epoch": 0.5109489051094891,
      "grad_norm": 0.6954646892322683,
      "learning_rate": 5e-06,
      "loss": 0.7693,
      "step": 140
    },
    {
      "epoch": 0.5474452554744526,
      "grad_norm": 0.7090301419858328,
      "learning_rate": 5e-06,
      "loss": 0.7698,
      "step": 150
    },
    {
      "epoch": 0.583941605839416,
      "grad_norm": 0.8080543613629317,
      "learning_rate": 5e-06,
      "loss": 0.7595,
      "step": 160
    },
    {
      "epoch": 0.6204379562043796,
      "grad_norm": 0.6419738319193917,
      "learning_rate": 5e-06,
      "loss": 0.7614,
      "step": 170
    },
    {
      "epoch": 0.656934306569343,
      "grad_norm": 0.5918340383797932,
      "learning_rate": 5e-06,
      "loss": 0.7541,
      "step": 180
    },
    {
      "epoch": 0.6934306569343066,
      "grad_norm": 1.152710341207945,
      "learning_rate": 5e-06,
      "loss": 0.7573,
      "step": 190
    },
    {
      "epoch": 0.7299270072992701,
      "grad_norm": 0.6391926393585599,
      "learning_rate": 5e-06,
      "loss": 0.7582,
      "step": 200
    },
    {
      "epoch": 0.7664233576642335,
      "grad_norm": 0.5988044301788384,
      "learning_rate": 5e-06,
      "loss": 0.7582,
      "step": 210
    },
    {
      "epoch": 0.8029197080291971,
      "grad_norm": 0.6020411724470022,
      "learning_rate": 5e-06,
      "loss": 0.7519,
      "step": 220
    },
    {
      "epoch": 0.8394160583941606,
      "grad_norm": 0.6220550687844844,
      "learning_rate": 5e-06,
      "loss": 0.7551,
      "step": 230
    },
    {
      "epoch": 0.8759124087591241,
      "grad_norm": 0.6082664111510806,
      "learning_rate": 5e-06,
      "loss": 0.7519,
      "step": 240
    },
    {
      "epoch": 0.9124087591240876,
      "grad_norm": 0.7929440801500307,
      "learning_rate": 5e-06,
      "loss": 0.7471,
      "step": 250
    },
    {
      "epoch": 0.948905109489051,
      "grad_norm": 0.8557482541404012,
      "learning_rate": 5e-06,
      "loss": 0.7471,
      "step": 260
    },
    {
      "epoch": 0.9854014598540146,
      "grad_norm": 0.5725008171409565,
      "learning_rate": 5e-06,
      "loss": 0.7448,
      "step": 270
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.7464940547943115,
      "eval_runtime": 193.5595,
      "eval_samples_per_second": 38.143,
      "eval_steps_per_second": 0.599,
      "step": 274
    },
    {
      "epoch": 1.0218978102189782,
      "grad_norm": 0.8391272846123186,
      "learning_rate": 5e-06,
      "loss": 0.7201,
      "step": 280
    },
    {
      "epoch": 1.0583941605839415,
      "grad_norm": 0.8538310326832037,
      "learning_rate": 5e-06,
      "loss": 0.6987,
      "step": 290
    },
    {
      "epoch": 1.094890510948905,
      "grad_norm": 0.6827445863139939,
      "learning_rate": 5e-06,
      "loss": 0.695,
      "step": 300
    },
    {
      "epoch": 1.1313868613138687,
      "grad_norm": 0.5882300886638387,
      "learning_rate": 5e-06,
      "loss": 0.6922,
      "step": 310
    },
    {
      "epoch": 1.167883211678832,
      "grad_norm": 0.9284544165624958,
      "learning_rate": 5e-06,
      "loss": 0.7005,
      "step": 320
    },
    {
      "epoch": 1.2043795620437956,
      "grad_norm": 0.7696038972626205,
      "learning_rate": 5e-06,
      "loss": 0.696,
      "step": 330
    },
    {
      "epoch": 1.2408759124087592,
      "grad_norm": 0.626331783091493,
      "learning_rate": 5e-06,
      "loss": 0.6961,
      "step": 340
    },
    {
      "epoch": 1.2773722627737225,
      "grad_norm": 0.6263624432151553,
      "learning_rate": 5e-06,
      "loss": 0.6955,
      "step": 350
    },
    {
      "epoch": 1.313868613138686,
      "grad_norm": 0.8006137682650242,
      "learning_rate": 5e-06,
      "loss": 0.6967,
      "step": 360
    },
    {
      "epoch": 1.3503649635036497,
      "grad_norm": 0.5880134638823885,
      "learning_rate": 5e-06,
      "loss": 0.702,
      "step": 370
    },
    {
      "epoch": 1.3868613138686132,
      "grad_norm": 0.647442922917404,
      "learning_rate": 5e-06,
      "loss": 0.7,
      "step": 380
    },
    {
      "epoch": 1.4233576642335766,
      "grad_norm": 0.6033654206417786,
      "learning_rate": 5e-06,
      "loss": 0.6935,
      "step": 390
    },
    {
      "epoch": 1.4598540145985401,
      "grad_norm": 0.842826594179501,
      "learning_rate": 5e-06,
      "loss": 0.6925,
      "step": 400
    },
    {
      "epoch": 1.4963503649635037,
      "grad_norm": 0.6487088398169627,
      "learning_rate": 5e-06,
      "loss": 0.6987,
      "step": 410
    },
    {
      "epoch": 1.5328467153284673,
      "grad_norm": 0.644019689180568,
      "learning_rate": 5e-06,
      "loss": 0.6975,
      "step": 420
    },
    {
      "epoch": 1.5693430656934306,
      "grad_norm": 0.7554928160990546,
      "learning_rate": 5e-06,
      "loss": 0.6955,
      "step": 430
    },
    {
      "epoch": 1.6058394160583942,
      "grad_norm": 0.6360155091593863,
      "learning_rate": 5e-06,
      "loss": 0.6947,
      "step": 440
    },
    {
      "epoch": 1.6423357664233578,
      "grad_norm": 0.7649060131558615,
      "learning_rate": 5e-06,
      "loss": 0.6979,
      "step": 450
    },
    {
      "epoch": 1.6788321167883211,
      "grad_norm": 0.7060249710510591,
      "learning_rate": 5e-06,
      "loss": 0.6887,
      "step": 460
    },
    {
      "epoch": 1.7153284671532847,
      "grad_norm": 0.614705856783836,
      "learning_rate": 5e-06,
      "loss": 0.6925,
      "step": 470
    },
    {
      "epoch": 1.7518248175182483,
      "grad_norm": 0.6665801961837532,
      "learning_rate": 5e-06,
      "loss": 0.6939,
      "step": 480
    },
    {
      "epoch": 1.7883211678832116,
      "grad_norm": 0.6137773686568165,
      "learning_rate": 5e-06,
      "loss": 0.6968,
      "step": 490
    },
    {
      "epoch": 1.8248175182481752,
      "grad_norm": 0.597915675609789,
      "learning_rate": 5e-06,
      "loss": 0.6903,
      "step": 500
    },
    {
      "epoch": 1.8613138686131387,
      "grad_norm": 0.5853039235207544,
      "learning_rate": 5e-06,
      "loss": 0.6919,
      "step": 510
    },
    {
      "epoch": 1.897810218978102,
      "grad_norm": 0.6252196584141081,
      "learning_rate": 5e-06,
      "loss": 0.6917,
      "step": 520
    },
    {
      "epoch": 1.9343065693430657,
      "grad_norm": 0.6355616628686108,
      "learning_rate": 5e-06,
      "loss": 0.6947,
      "step": 530
    },
    {
      "epoch": 1.9708029197080292,
      "grad_norm": 0.7084920207276291,
      "learning_rate": 5e-06,
      "loss": 0.6909,
      "step": 540
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.7327279448509216,
      "eval_runtime": 184.0166,
      "eval_samples_per_second": 40.121,
      "eval_steps_per_second": 0.63,
      "step": 548
    },
    {
      "epoch": 2.0072992700729926,
      "grad_norm": 0.8657992423171681,
      "learning_rate": 5e-06,
      "loss": 0.6891,
      "step": 550
    },
    {
      "epoch": 2.0437956204379564,
      "grad_norm": 0.7941631908501828,
      "learning_rate": 5e-06,
      "loss": 0.6428,
      "step": 560
    },
    {
      "epoch": 2.0802919708029197,
      "grad_norm": 0.9214323090929812,
      "learning_rate": 5e-06,
      "loss": 0.6418,
      "step": 570
    },
    {
      "epoch": 2.116788321167883,
      "grad_norm": 0.7759251168034266,
      "learning_rate": 5e-06,
      "loss": 0.6411,
      "step": 580
    },
    {
      "epoch": 2.153284671532847,
      "grad_norm": 0.7159794483510337,
      "learning_rate": 5e-06,
      "loss": 0.6429,
      "step": 590
    },
    {
      "epoch": 2.18978102189781,
      "grad_norm": 0.7275049985326262,
      "learning_rate": 5e-06,
      "loss": 0.6418,
      "step": 600
    },
    {
      "epoch": 2.2262773722627736,
      "grad_norm": 0.5798767855808342,
      "learning_rate": 5e-06,
      "loss": 0.6455,
      "step": 610
    },
    {
      "epoch": 2.2627737226277373,
      "grad_norm": 0.5604983214077892,
      "learning_rate": 5e-06,
      "loss": 0.64,
      "step": 620
    },
    {
      "epoch": 2.2992700729927007,
      "grad_norm": 0.6059873724806449,
      "learning_rate": 5e-06,
      "loss": 0.6424,
      "step": 630
    },
    {
      "epoch": 2.335766423357664,
      "grad_norm": 0.8067784530621691,
      "learning_rate": 5e-06,
      "loss": 0.6422,
      "step": 640
    },
    {
      "epoch": 2.372262773722628,
      "grad_norm": 0.7136802882114688,
      "learning_rate": 5e-06,
      "loss": 0.6437,
      "step": 650
    },
    {
      "epoch": 2.408759124087591,
      "grad_norm": 0.6102999029962267,
      "learning_rate": 5e-06,
      "loss": 0.6458,
      "step": 660
    },
    {
      "epoch": 2.445255474452555,
      "grad_norm": 0.614293225765085,
      "learning_rate": 5e-06,
      "loss": 0.6452,
      "step": 670
    },
    {
      "epoch": 2.4817518248175183,
      "grad_norm": 0.600198323252072,
      "learning_rate": 5e-06,
      "loss": 0.644,
      "step": 680
    },
    {
      "epoch": 2.5182481751824817,
      "grad_norm": 0.7644298597279312,
      "learning_rate": 5e-06,
      "loss": 0.6491,
      "step": 690
    },
    {
      "epoch": 2.554744525547445,
      "grad_norm": 0.778090825027965,
      "learning_rate": 5e-06,
      "loss": 0.6431,
      "step": 700
    },
    {
      "epoch": 2.591240875912409,
      "grad_norm": 0.6476324319228641,
      "learning_rate": 5e-06,
      "loss": 0.6473,
      "step": 710
    },
    {
      "epoch": 2.627737226277372,
      "grad_norm": 0.5998032151877339,
      "learning_rate": 5e-06,
      "loss": 0.6465,
      "step": 720
    },
    {
      "epoch": 2.664233576642336,
      "grad_norm": 0.5955651290954712,
      "learning_rate": 5e-06,
      "loss": 0.6456,
      "step": 730
    },
    {
      "epoch": 2.7007299270072993,
      "grad_norm": 0.6010226042821943,
      "learning_rate": 5e-06,
      "loss": 0.6442,
      "step": 740
    },
    {
      "epoch": 2.7372262773722627,
      "grad_norm": 0.821931399421341,
      "learning_rate": 5e-06,
      "loss": 0.6459,
      "step": 750
    },
    {
      "epoch": 2.7737226277372264,
      "grad_norm": 0.6668282736525819,
      "learning_rate": 5e-06,
      "loss": 0.6503,
      "step": 760
    },
    {
      "epoch": 2.81021897810219,
      "grad_norm": 0.7099179596842483,
      "learning_rate": 5e-06,
      "loss": 0.6473,
      "step": 770
    },
    {
      "epoch": 2.846715328467153,
      "grad_norm": 0.6671120507448849,
      "learning_rate": 5e-06,
      "loss": 0.6426,
      "step": 780
    },
    {
      "epoch": 2.883211678832117,
      "grad_norm": 0.6360799683340312,
      "learning_rate": 5e-06,
      "loss": 0.6473,
      "step": 790
    },
    {
      "epoch": 2.9197080291970803,
      "grad_norm": 0.7340811126545655,
      "learning_rate": 5e-06,
      "loss": 0.651,
      "step": 800
    },
    {
      "epoch": 2.9562043795620436,
      "grad_norm": 0.618110632863654,
      "learning_rate": 5e-06,
      "loss": 0.6448,
      "step": 810
    },
    {
      "epoch": 2.9927007299270074,
      "grad_norm": 0.7521256358398457,
      "learning_rate": 5e-06,
      "loss": 0.6423,
      "step": 820
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.7341417074203491,
      "eval_runtime": 183.6218,
      "eval_samples_per_second": 40.208,
      "eval_steps_per_second": 0.632,
      "step": 822
    },
    {
      "epoch": 3.0,
      "step": 822,
      "total_flos": 1376671236096000.0,
      "train_loss": 0.7106172601962032,
      "train_runtime": 27349.7837,
      "train_samples_per_second": 15.387,
      "train_steps_per_second": 0.03
    }
  ],
  "logging_steps": 10,
  "max_steps": 822,
  "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": 1376671236096000.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}