{
  "best_metric": 0.06379189342260361,
  "best_model_checkpoint": "./beans_outputs/checkpoint-520",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 650,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07692307692307693,
      "grad_norm": 2.1982851028442383,
      "learning_rate": 1.9692307692307696e-05,
      "loss": 1.0245,
      "step": 10
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 1.858752965927124,
      "learning_rate": 1.9384615384615386e-05,
      "loss": 0.9453,
      "step": 20
    },
    {
      "epoch": 0.23076923076923078,
      "grad_norm": 2.0879430770874023,
      "learning_rate": 1.907692307692308e-05,
      "loss": 0.8405,
      "step": 30
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 2.6220877170562744,
      "learning_rate": 1.876923076923077e-05,
      "loss": 0.6952,
      "step": 40
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 3.9979019165039062,
      "learning_rate": 1.8461538461538465e-05,
      "loss": 0.6624,
      "step": 50
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 2.129120349884033,
      "learning_rate": 1.8153846153846155e-05,
      "loss": 0.587,
      "step": 60
    },
    {
      "epoch": 0.5384615384615384,
      "grad_norm": 2.488262414932251,
      "learning_rate": 1.784615384615385e-05,
      "loss": 0.5232,
      "step": 70
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 1.234305500984192,
      "learning_rate": 1.753846153846154e-05,
      "loss": 0.3995,
      "step": 80
    },
    {
      "epoch": 0.6923076923076923,
      "grad_norm": 2.4971282482147217,
      "learning_rate": 1.7230769230769234e-05,
      "loss": 0.357,
      "step": 90
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 3.0826847553253174,
      "learning_rate": 1.6923076923076924e-05,
      "loss": 0.3739,
      "step": 100
    },
    {
      "epoch": 0.8461538461538461,
      "grad_norm": 2.1428916454315186,
      "learning_rate": 1.6615384615384618e-05,
      "loss": 0.3331,
      "step": 110
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 2.439236879348755,
      "learning_rate": 1.630769230769231e-05,
      "loss": 0.2457,
      "step": 120
    },
    {
      "epoch": 1.0,
      "grad_norm": 6.936716079711914,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.2816,
      "step": 130
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9699248120300752,
      "eval_loss": 0.2115139216184616,
      "eval_runtime": 2.5325,
      "eval_samples_per_second": 52.516,
      "eval_steps_per_second": 6.713,
      "step": 130
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 1.6164023876190186,
      "learning_rate": 1.5692307692307693e-05,
      "loss": 0.1984,
      "step": 140
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 3.183772563934326,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.2249,
      "step": 150
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 1.4262399673461914,
      "learning_rate": 1.5076923076923078e-05,
      "loss": 0.2014,
      "step": 160
    },
    {
      "epoch": 1.3076923076923077,
      "grad_norm": 0.5211865901947021,
      "learning_rate": 1.4769230769230772e-05,
      "loss": 0.2194,
      "step": 170
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.5973725914955139,
      "learning_rate": 1.4461538461538462e-05,
      "loss": 0.2049,
      "step": 180
    },
    {
      "epoch": 1.4615384615384617,
      "grad_norm": 3.3427956104278564,
      "learning_rate": 1.4153846153846156e-05,
      "loss": 0.1615,
      "step": 190
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.45850175619125366,
      "learning_rate": 1.3846153846153847e-05,
      "loss": 0.2527,
      "step": 200
    },
    {
      "epoch": 1.6153846153846154,
      "grad_norm": 0.634377121925354,
      "learning_rate": 1.353846153846154e-05,
      "loss": 0.1794,
      "step": 210
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 1.9677997827529907,
      "learning_rate": 1.3230769230769231e-05,
      "loss": 0.2115,
      "step": 220
    },
    {
      "epoch": 1.7692307692307692,
      "grad_norm": 1.920150876045227,
      "learning_rate": 1.2923076923076925e-05,
      "loss": 0.1757,
      "step": 230
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 5.09351110458374,
      "learning_rate": 1.2615384615384616e-05,
      "loss": 0.2563,
      "step": 240
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 0.3909279406070709,
      "learning_rate": 1.230769230769231e-05,
      "loss": 0.1386,
      "step": 250
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.34947332739830017,
      "learning_rate": 1.2e-05,
      "loss": 0.1323,
      "step": 260
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.1257385015487671,
      "eval_runtime": 2.1658,
      "eval_samples_per_second": 61.408,
      "eval_steps_per_second": 7.849,
      "step": 260
    },
    {
      "epoch": 2.076923076923077,
      "grad_norm": 0.5078479051589966,
      "learning_rate": 1.1692307692307694e-05,
      "loss": 0.0898,
      "step": 270
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.3371464014053345,
      "learning_rate": 1.1384615384615385e-05,
      "loss": 0.1393,
      "step": 280
    },
    {
      "epoch": 2.230769230769231,
      "grad_norm": 7.1395697593688965,
      "learning_rate": 1.1076923076923079e-05,
      "loss": 0.128,
      "step": 290
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 7.69725227355957,
      "learning_rate": 1.076923076923077e-05,
      "loss": 0.1366,
      "step": 300
    },
    {
      "epoch": 2.3846153846153846,
      "grad_norm": 6.024722576141357,
      "learning_rate": 1.0461538461538463e-05,
      "loss": 0.1756,
      "step": 310
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.3003959059715271,
      "learning_rate": 1.0153846153846154e-05,
      "loss": 0.1551,
      "step": 320
    },
    {
      "epoch": 2.5384615384615383,
      "grad_norm": 6.158186435699463,
      "learning_rate": 9.846153846153848e-06,
      "loss": 0.1849,
      "step": 330
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 1.280966877937317,
      "learning_rate": 9.53846153846154e-06,
      "loss": 0.0763,
      "step": 340
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 0.34941595792770386,
      "learning_rate": 9.230769230769232e-06,
      "loss": 0.1369,
      "step": 350
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 7.8951897621154785,
      "learning_rate": 8.923076923076925e-06,
      "loss": 0.1189,
      "step": 360
    },
    {
      "epoch": 2.8461538461538463,
      "grad_norm": 0.3557591438293457,
      "learning_rate": 8.615384615384617e-06,
      "loss": 0.1519,
      "step": 370
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 1.2551947832107544,
      "learning_rate": 8.307692307692309e-06,
      "loss": 0.1318,
      "step": 380
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.3029463291168213,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.1416,
      "step": 390
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.09356901794672012,
      "eval_runtime": 2.0719,
      "eval_samples_per_second": 64.192,
      "eval_steps_per_second": 8.205,
      "step": 390
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 2.313016414642334,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.1328,
      "step": 400
    },
    {
      "epoch": 3.1538461538461537,
      "grad_norm": 6.524436950683594,
      "learning_rate": 7.384615384615386e-06,
      "loss": 0.1526,
      "step": 410
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 8.927879333496094,
      "learning_rate": 7.076923076923078e-06,
      "loss": 0.1469,
      "step": 420
    },
    {
      "epoch": 3.3076923076923075,
      "grad_norm": 0.23213094472885132,
      "learning_rate": 6.76923076923077e-06,
      "loss": 0.0765,
      "step": 430
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 8.142635345458984,
      "learning_rate": 6.461538461538463e-06,
      "loss": 0.1189,
      "step": 440
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 12.428342819213867,
      "learning_rate": 6.153846153846155e-06,
      "loss": 0.0813,
      "step": 450
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 4.8793134689331055,
      "learning_rate": 5.846153846153847e-06,
      "loss": 0.0968,
      "step": 460
    },
    {
      "epoch": 3.6153846153846154,
      "grad_norm": 8.63253402709961,
      "learning_rate": 5.538461538461539e-06,
      "loss": 0.0781,
      "step": 470
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.22319431602954865,
      "learning_rate": 5.230769230769232e-06,
      "loss": 0.0955,
      "step": 480
    },
    {
      "epoch": 3.769230769230769,
      "grad_norm": 0.2103033810853958,
      "learning_rate": 4.923076923076924e-06,
      "loss": 0.0781,
      "step": 490
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.9193207025527954,
      "learning_rate": 4.615384615384616e-06,
      "loss": 0.1432,
      "step": 500
    },
    {
      "epoch": 3.9230769230769234,
      "grad_norm": 0.20915640890598297,
      "learning_rate": 4.307692307692308e-06,
      "loss": 0.1354,
      "step": 510
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.3006608784198761,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0855,
      "step": 520
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9924812030075187,
      "eval_loss": 0.06379189342260361,
      "eval_runtime": 2.378,
      "eval_samples_per_second": 55.93,
      "eval_steps_per_second": 7.149,
      "step": 520
    },
    {
      "epoch": 4.076923076923077,
      "grad_norm": 6.645312309265137,
      "learning_rate": 3.692307692307693e-06,
      "loss": 0.0971,
      "step": 530
    },
    {
      "epoch": 4.153846153846154,
      "grad_norm": 1.5618813037872314,
      "learning_rate": 3.384615384615385e-06,
      "loss": 0.1609,
      "step": 540
    },
    {
      "epoch": 4.230769230769231,
      "grad_norm": 0.3374852240085602,
      "learning_rate": 3.0769230769230774e-06,
      "loss": 0.1527,
      "step": 550
    },
    {
      "epoch": 4.3076923076923075,
      "grad_norm": 1.1344232559204102,
      "learning_rate": 2.7692307692307697e-06,
      "loss": 0.0499,
      "step": 560
    },
    {
      "epoch": 4.384615384615385,
      "grad_norm": 0.3350333869457245,
      "learning_rate": 2.461538461538462e-06,
      "loss": 0.1345,
      "step": 570
    },
    {
      "epoch": 4.461538461538462,
      "grad_norm": 0.271489679813385,
      "learning_rate": 2.153846153846154e-06,
      "loss": 0.0877,
      "step": 580
    },
    {
      "epoch": 4.538461538461538,
      "grad_norm": 8.799779891967773,
      "learning_rate": 1.8461538461538465e-06,
      "loss": 0.0742,
      "step": 590
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 2.9175331592559814,
      "learning_rate": 1.5384615384615387e-06,
      "loss": 0.0933,
      "step": 600
    },
    {
      "epoch": 4.6923076923076925,
      "grad_norm": 5.1075568199157715,
      "learning_rate": 1.230769230769231e-06,
      "loss": 0.0854,
      "step": 610
    },
    {
      "epoch": 4.769230769230769,
      "grad_norm": 0.25091925263404846,
      "learning_rate": 9.230769230769232e-07,
      "loss": 0.0866,
      "step": 620
    },
    {
      "epoch": 4.846153846153846,
      "grad_norm": 0.22164465487003326,
      "learning_rate": 6.153846153846155e-07,
      "loss": 0.0716,
      "step": 630
    },
    {
      "epoch": 4.923076923076923,
      "grad_norm": 0.22948431968688965,
      "learning_rate": 3.0769230769230774e-07,
      "loss": 0.0765,
      "step": 640
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.59376460313797,
      "learning_rate": 0.0,
      "loss": 0.1177,
      "step": 650
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9849624060150376,
      "eval_loss": 0.07658705860376358,
      "eval_runtime": 2.1789,
      "eval_samples_per_second": 61.04,
      "eval_steps_per_second": 7.802,
      "step": 650
    },
    {
      "epoch": 5.0,
      "step": 650,
      "total_flos": 4.006371770595533e+17,
      "train_loss": 0.22003128262666555,
      "train_runtime": 354.3539,
      "train_samples_per_second": 14.59,
      "train_steps_per_second": 1.834
    }
  ],
  "logging_steps": 10,
  "max_steps": 650,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.006371770595533e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}