{
  "best_global_step": 705,
  "best_metric": 0.16024889051914215,
  "best_model_checkpoint": "./anime_out/checkpoint-705",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 705,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07092198581560284,
      "grad_norm": 1956.7347412109375,
      "learning_rate": 1.9858156028368796e-05,
      "loss": 2.5378,
      "step": 10
    },
    {
      "epoch": 0.14184397163120568,
      "grad_norm": 398.7348327636719,
      "learning_rate": 1.971631205673759e-05,
      "loss": 0.7439,
      "step": 20
    },
    {
      "epoch": 0.2127659574468085,
      "grad_norm": 267.0840759277344,
      "learning_rate": 1.9574468085106384e-05,
      "loss": 1.8382,
      "step": 30
    },
    {
      "epoch": 0.28368794326241137,
      "grad_norm": 123.07373809814453,
      "learning_rate": 1.9432624113475178e-05,
      "loss": 2.3966,
      "step": 40
    },
    {
      "epoch": 0.3546099290780142,
      "grad_norm": 8.433001494267955e-05,
      "learning_rate": 1.929078014184397e-05,
      "loss": 1.4528,
      "step": 50
    },
    {
      "epoch": 0.425531914893617,
      "grad_norm": 310.42840576171875,
      "learning_rate": 1.914893617021277e-05,
      "loss": 1.5857,
      "step": 60
    },
    {
      "epoch": 0.49645390070921985,
      "grad_norm": 14.845270156860352,
      "learning_rate": 1.9007092198581563e-05,
      "loss": 0.6746,
      "step": 70
    },
    {
      "epoch": 0.5673758865248227,
      "grad_norm": 355.8478698730469,
      "learning_rate": 1.8865248226950357e-05,
      "loss": 1.2875,
      "step": 80
    },
    {
      "epoch": 0.6382978723404256,
      "grad_norm": 571.3264770507812,
      "learning_rate": 1.872340425531915e-05,
      "loss": 0.4645,
      "step": 90
    },
    {
      "epoch": 0.7092198581560284,
      "grad_norm": 270.5877380371094,
      "learning_rate": 1.8581560283687945e-05,
      "loss": 2.0277,
      "step": 100
    },
    {
      "epoch": 0.7801418439716312,
      "grad_norm": 14.781210899353027,
      "learning_rate": 1.843971631205674e-05,
      "loss": 1.1098,
      "step": 110
    },
    {
      "epoch": 0.851063829787234,
      "grad_norm": 323.6094665527344,
      "learning_rate": 1.8297872340425533e-05,
      "loss": 1.0614,
      "step": 120
    },
    {
      "epoch": 0.9219858156028369,
      "grad_norm": 469.43707275390625,
      "learning_rate": 1.8156028368794327e-05,
      "loss": 0.8355,
      "step": 130
    },
    {
      "epoch": 0.9929078014184397,
      "grad_norm": 0.007668596692383289,
      "learning_rate": 1.801418439716312e-05,
      "loss": 0.5349,
      "step": 140
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9197860962566845,
      "eval_loss": 0.36517199873924255,
      "eval_runtime": 19.748,
      "eval_samples_per_second": 9.469,
      "eval_steps_per_second": 1.215,
      "step": 141
    },
    {
      "epoch": 1.0638297872340425,
      "grad_norm": 1.1018470525741577,
      "learning_rate": 1.7872340425531915e-05,
      "loss": 0.4347,
      "step": 150
    },
    {
      "epoch": 1.1347517730496455,
      "grad_norm": 2.6226277351379395,
      "learning_rate": 1.773049645390071e-05,
      "loss": 0.6987,
      "step": 160
    },
    {
      "epoch": 1.2056737588652482,
      "grad_norm": 0.04894453287124634,
      "learning_rate": 1.7588652482269506e-05,
      "loss": 0.7401,
      "step": 170
    },
    {
      "epoch": 1.2765957446808511,
      "grad_norm": 0.19339050352573395,
      "learning_rate": 1.74468085106383e-05,
      "loss": 0.5764,
      "step": 180
    },
    {
      "epoch": 1.3475177304964538,
      "grad_norm": 176.8428192138672,
      "learning_rate": 1.7304964539007094e-05,
      "loss": 0.4433,
      "step": 190
    },
    {
      "epoch": 1.4184397163120568,
      "grad_norm": 253.8905029296875,
      "learning_rate": 1.716312056737589e-05,
      "loss": 0.5627,
      "step": 200
    },
    {
      "epoch": 1.4893617021276595,
      "grad_norm": 191.34634399414062,
      "learning_rate": 1.7021276595744682e-05,
      "loss": 0.757,
      "step": 210
    },
    {
      "epoch": 1.5602836879432624,
      "grad_norm": 436.973876953125,
      "learning_rate": 1.6879432624113476e-05,
      "loss": 1.9875,
      "step": 220
    },
    {
      "epoch": 1.6312056737588652,
      "grad_norm": 412.5374755859375,
      "learning_rate": 1.673758865248227e-05,
      "loss": 0.8893,
      "step": 230
    },
    {
      "epoch": 1.702127659574468,
      "grad_norm": 153.39613342285156,
      "learning_rate": 1.6595744680851064e-05,
      "loss": 1.1219,
      "step": 240
    },
    {
      "epoch": 1.773049645390071,
      "grad_norm": 303.0024108886719,
      "learning_rate": 1.645390070921986e-05,
      "loss": 0.8197,
      "step": 250
    },
    {
      "epoch": 1.8439716312056738,
      "grad_norm": 4.597931385040283,
      "learning_rate": 1.6312056737588652e-05,
      "loss": 0.3096,
      "step": 260
    },
    {
      "epoch": 1.9148936170212765,
      "grad_norm": 7.166381359100342,
      "learning_rate": 1.6170212765957446e-05,
      "loss": 0.304,
      "step": 270
    },
    {
      "epoch": 1.9858156028368794,
      "grad_norm": 301.7471618652344,
      "learning_rate": 1.6028368794326244e-05,
      "loss": 0.6178,
      "step": 280
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9251336898395722,
      "eval_loss": 0.28867608308792114,
      "eval_runtime": 19.4186,
      "eval_samples_per_second": 9.63,
      "eval_steps_per_second": 1.236,
      "step": 282
    },
    {
      "epoch": 2.0567375886524824,
      "grad_norm": 2.6156604290008545,
      "learning_rate": 1.5886524822695038e-05,
      "loss": 0.2282,
      "step": 290
    },
    {
      "epoch": 2.127659574468085,
      "grad_norm": 0.07026529312133789,
      "learning_rate": 1.5744680851063832e-05,
      "loss": 0.2304,
      "step": 300
    },
    {
      "epoch": 2.198581560283688,
      "grad_norm": 0.2403770089149475,
      "learning_rate": 1.5602836879432626e-05,
      "loss": 0.1619,
      "step": 310
    },
    {
      "epoch": 2.269503546099291,
      "grad_norm": 1001.3410034179688,
      "learning_rate": 1.546099290780142e-05,
      "loss": 0.4448,
      "step": 320
    },
    {
      "epoch": 2.3404255319148937,
      "grad_norm": 0.13391649723052979,
      "learning_rate": 1.5319148936170214e-05,
      "loss": 0.2728,
      "step": 330
    },
    {
      "epoch": 2.4113475177304964,
      "grad_norm": 189.5388946533203,
      "learning_rate": 1.5177304964539008e-05,
      "loss": 0.7779,
      "step": 340
    },
    {
      "epoch": 2.482269503546099,
      "grad_norm": 1.5827876040930278e-06,
      "learning_rate": 1.5035460992907802e-05,
      "loss": 0.8039,
      "step": 350
    },
    {
      "epoch": 2.5531914893617023,
      "grad_norm": 127.49388885498047,
      "learning_rate": 1.4893617021276596e-05,
      "loss": 0.6087,
      "step": 360
    },
    {
      "epoch": 2.624113475177305,
      "grad_norm": 0.00041575109935365617,
      "learning_rate": 1.475177304964539e-05,
      "loss": 0.468,
      "step": 370
    },
    {
      "epoch": 2.6950354609929077,
      "grad_norm": 637.6605224609375,
      "learning_rate": 1.4609929078014187e-05,
      "loss": 0.4797,
      "step": 380
    },
    {
      "epoch": 2.7659574468085104,
      "grad_norm": 0.000833332072943449,
      "learning_rate": 1.4468085106382981e-05,
      "loss": 0.7188,
      "step": 390
    },
    {
      "epoch": 2.8368794326241136,
      "grad_norm": 0.004562568850815296,
      "learning_rate": 1.4326241134751775e-05,
      "loss": 0.1451,
      "step": 400
    },
    {
      "epoch": 2.9078014184397163,
      "grad_norm": 360.48321533203125,
      "learning_rate": 1.418439716312057e-05,
      "loss": 0.7191,
      "step": 410
    },
    {
      "epoch": 2.978723404255319,
      "grad_norm": 236.08912658691406,
      "learning_rate": 1.4042553191489363e-05,
      "loss": 0.3556,
      "step": 420
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9197860962566845,
      "eval_loss": 0.297281414270401,
      "eval_runtime": 19.4814,
      "eval_samples_per_second": 9.599,
      "eval_steps_per_second": 1.232,
      "step": 423
    },
    {
      "epoch": 3.049645390070922,
      "grad_norm": 222.08946228027344,
      "learning_rate": 1.3900709219858157e-05,
      "loss": 0.6342,
      "step": 430
    },
    {
      "epoch": 3.120567375886525,
      "grad_norm": 0.12243126332759857,
      "learning_rate": 1.3758865248226951e-05,
      "loss": 0.9937,
      "step": 440
    },
    {
      "epoch": 3.1914893617021276,
      "grad_norm": 24.167377471923828,
      "learning_rate": 1.3617021276595745e-05,
      "loss": 0.122,
      "step": 450
    },
    {
      "epoch": 3.2624113475177303,
      "grad_norm": 0.006283226888626814,
      "learning_rate": 1.347517730496454e-05,
      "loss": 0.2216,
      "step": 460
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 41.654052734375,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.0438,
      "step": 470
    },
    {
      "epoch": 3.404255319148936,
      "grad_norm": 317.71917724609375,
      "learning_rate": 1.3191489361702127e-05,
      "loss": 0.7924,
      "step": 480
    },
    {
      "epoch": 3.475177304964539,
      "grad_norm": 1.5460463762283325,
      "learning_rate": 1.3049645390070925e-05,
      "loss": 0.0088,
      "step": 490
    },
    {
      "epoch": 3.546099290780142,
      "grad_norm": 121.88572692871094,
      "learning_rate": 1.2907801418439719e-05,
      "loss": 0.2176,
      "step": 500
    },
    {
      "epoch": 3.617021276595745,
      "grad_norm": 0.06358540803194046,
      "learning_rate": 1.2765957446808513e-05,
      "loss": 0.1413,
      "step": 510
    },
    {
      "epoch": 3.6879432624113475,
      "grad_norm": 147.4714813232422,
      "learning_rate": 1.2624113475177307e-05,
      "loss": 0.1081,
      "step": 520
    },
    {
      "epoch": 3.7588652482269502,
      "grad_norm": 0.18462225794792175,
      "learning_rate": 1.24822695035461e-05,
      "loss": 0.1447,
      "step": 530
    },
    {
      "epoch": 3.829787234042553,
      "grad_norm": 0.011159995570778847,
      "learning_rate": 1.2340425531914895e-05,
      "loss": 0.3039,
      "step": 540
    },
    {
      "epoch": 3.900709219858156,
      "grad_norm": 5.591824054718018,
      "learning_rate": 1.2198581560283689e-05,
      "loss": 0.3814,
      "step": 550
    },
    {
      "epoch": 3.971631205673759,
      "grad_norm": 0.0026896377094089985,
      "learning_rate": 1.2056737588652483e-05,
      "loss": 0.2332,
      "step": 560
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9518716577540107,
      "eval_loss": 0.20226924121379852,
      "eval_runtime": 19.4299,
      "eval_samples_per_second": 9.624,
      "eval_steps_per_second": 1.235,
      "step": 564
    },
    {
      "epoch": 4.042553191489362,
      "grad_norm": 49.80411911010742,
      "learning_rate": 1.1914893617021277e-05,
      "loss": 1.1912,
      "step": 570
    },
    {
      "epoch": 4.113475177304965,
      "grad_norm": 485.7699279785156,
      "learning_rate": 1.177304964539007e-05,
      "loss": 0.5817,
      "step": 580
    },
    {
      "epoch": 4.184397163120567,
      "grad_norm": 19.611454010009766,
      "learning_rate": 1.1631205673758865e-05,
      "loss": 0.1125,
      "step": 590
    },
    {
      "epoch": 4.25531914893617,
      "grad_norm": 36.56155014038086,
      "learning_rate": 1.1489361702127662e-05,
      "loss": 0.6281,
      "step": 600
    },
    {
      "epoch": 4.326241134751773,
      "grad_norm": 279.4352722167969,
      "learning_rate": 1.1347517730496456e-05,
      "loss": 0.9411,
      "step": 610
    },
    {
      "epoch": 4.397163120567376,
      "grad_norm": 8.183698654174805,
      "learning_rate": 1.120567375886525e-05,
      "loss": 0.0053,
      "step": 620
    },
    {
      "epoch": 4.468085106382979,
      "grad_norm": 0.005647369660437107,
      "learning_rate": 1.1063829787234044e-05,
      "loss": 0.6476,
      "step": 630
    },
    {
      "epoch": 4.539007092198582,
      "grad_norm": 76.17412567138672,
      "learning_rate": 1.0921985815602838e-05,
      "loss": 0.2671,
      "step": 640
    },
    {
      "epoch": 4.609929078014185,
      "grad_norm": 177.41612243652344,
      "learning_rate": 1.0780141843971632e-05,
      "loss": 0.3544,
      "step": 650
    },
    {
      "epoch": 4.680851063829787,
      "grad_norm": 304.32763671875,
      "learning_rate": 1.0638297872340426e-05,
      "loss": 0.6,
      "step": 660
    },
    {
      "epoch": 4.75177304964539,
      "grad_norm": 1.9934466763515957e-05,
      "learning_rate": 1.049645390070922e-05,
      "loss": 0.1934,
      "step": 670
    },
    {
      "epoch": 4.822695035460993,
      "grad_norm": 284.1419677734375,
      "learning_rate": 1.0354609929078014e-05,
      "loss": 0.5667,
      "step": 680
    },
    {
      "epoch": 4.8936170212765955,
      "grad_norm": 12.019274711608887,
      "learning_rate": 1.0212765957446808e-05,
      "loss": 0.4064,
      "step": 690
    },
    {
      "epoch": 4.964539007092198,
      "grad_norm": 0.014889250509440899,
      "learning_rate": 1.0070921985815602e-05,
      "loss": 0.4825,
      "step": 700
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9625668449197861,
      "eval_loss": 0.16024889051914215,
      "eval_runtime": 19.4874,
      "eval_samples_per_second": 9.596,
      "eval_steps_per_second": 1.232,
      "step": 705
    }
  ],
  "logging_steps": 10,
  "max_steps": 1410,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.0811885244459827e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}