{
  "best_metric": 0.06696975976228714,
  "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": 2.1988484859466553,
      "learning_rate": 1.9692307692307696e-05,
      "loss": 1.0245,
      "step": 10
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 2.0091118812561035,
      "learning_rate": 1.9384615384615386e-05,
      "loss": 0.9455,
      "step": 20
    },
    {
      "epoch": 0.23076923076923078,
      "grad_norm": 2.0441346168518066,
      "learning_rate": 1.907692307692308e-05,
      "loss": 0.84,
      "step": 30
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 2.614025354385376,
      "learning_rate": 1.876923076923077e-05,
      "loss": 0.6962,
      "step": 40
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 4.7578816413879395,
      "learning_rate": 1.8461538461538465e-05,
      "loss": 0.6628,
      "step": 50
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 2.082709550857544,
      "learning_rate": 1.8153846153846155e-05,
      "loss": 0.5827,
      "step": 60
    },
    {
      "epoch": 0.5384615384615384,
      "grad_norm": 2.5298991203308105,
      "learning_rate": 1.784615384615385e-05,
      "loss": 0.5206,
      "step": 70
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 1.2223843336105347,
      "learning_rate": 1.753846153846154e-05,
      "loss": 0.3961,
      "step": 80
    },
    {
      "epoch": 0.6923076923076923,
      "grad_norm": 2.384169816970825,
      "learning_rate": 1.7230769230769234e-05,
      "loss": 0.3538,
      "step": 90
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 2.968139171600342,
      "learning_rate": 1.6923076923076924e-05,
      "loss": 0.366,
      "step": 100
    },
    {
      "epoch": 0.8461538461538461,
      "grad_norm": 2.683225393295288,
      "learning_rate": 1.6615384615384618e-05,
      "loss": 0.332,
      "step": 110
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 2.5319671630859375,
      "learning_rate": 1.630769230769231e-05,
      "loss": 0.2441,
      "step": 120
    },
    {
      "epoch": 1.0,
      "grad_norm": 7.455262660980225,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.2797,
      "step": 130
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9624060150375939,
      "eval_loss": 0.21510010957717896,
      "eval_runtime": 0.6622,
      "eval_samples_per_second": 200.836,
      "eval_steps_per_second": 25.671,
      "step": 130
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 1.613183856010437,
      "learning_rate": 1.5692307692307693e-05,
      "loss": 0.1994,
      "step": 140
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 3.0574400424957275,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.2208,
      "step": 150
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 1.4466005563735962,
      "learning_rate": 1.5076923076923078e-05,
      "loss": 0.2071,
      "step": 160
    },
    {
      "epoch": 1.3076923076923077,
      "grad_norm": 0.5133665204048157,
      "learning_rate": 1.4769230769230772e-05,
      "loss": 0.2195,
      "step": 170
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.5961588621139526,
      "learning_rate": 1.4461538461538462e-05,
      "loss": 0.2082,
      "step": 180
    },
    {
      "epoch": 1.4615384615384617,
      "grad_norm": 3.2639622688293457,
      "learning_rate": 1.4153846153846156e-05,
      "loss": 0.1641,
      "step": 190
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.46516209840774536,
      "learning_rate": 1.3846153846153847e-05,
      "loss": 0.2525,
      "step": 200
    },
    {
      "epoch": 1.6153846153846154,
      "grad_norm": 0.6015739440917969,
      "learning_rate": 1.353846153846154e-05,
      "loss": 0.1814,
      "step": 210
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 3.2524654865264893,
      "learning_rate": 1.3230769230769231e-05,
      "loss": 0.2178,
      "step": 220
    },
    {
      "epoch": 1.7692307692307692,
      "grad_norm": 2.14794659614563,
      "learning_rate": 1.2923076923076925e-05,
      "loss": 0.1855,
      "step": 230
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 5.231557369232178,
      "learning_rate": 1.2615384615384616e-05,
      "loss": 0.2585,
      "step": 240
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 0.37082186341285706,
      "learning_rate": 1.230769230769231e-05,
      "loss": 0.1372,
      "step": 250
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.3476381003856659,
      "learning_rate": 1.2e-05,
      "loss": 0.1301,
      "step": 260
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.12592361867427826,
      "eval_runtime": 0.5663,
      "eval_samples_per_second": 234.86,
      "eval_steps_per_second": 30.02,
      "step": 260
    },
    {
      "epoch": 2.076923076923077,
      "grad_norm": 0.48094871640205383,
      "learning_rate": 1.1692307692307694e-05,
      "loss": 0.0947,
      "step": 270
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.34189698100090027,
      "learning_rate": 1.1384615384615385e-05,
      "loss": 0.1412,
      "step": 280
    },
    {
      "epoch": 2.230769230769231,
      "grad_norm": 7.666577339172363,
      "learning_rate": 1.1076923076923079e-05,
      "loss": 0.1267,
      "step": 290
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 8.487817764282227,
      "learning_rate": 1.076923076923077e-05,
      "loss": 0.1272,
      "step": 300
    },
    {
      "epoch": 2.3846153846153846,
      "grad_norm": 4.255734443664551,
      "learning_rate": 1.0461538461538463e-05,
      "loss": 0.1767,
      "step": 310
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.30839142203330994,
      "learning_rate": 1.0153846153846154e-05,
      "loss": 0.1465,
      "step": 320
    },
    {
      "epoch": 2.5384615384615383,
      "grad_norm": 5.301502704620361,
      "learning_rate": 9.846153846153848e-06,
      "loss": 0.1855,
      "step": 330
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 1.1118645668029785,
      "learning_rate": 9.53846153846154e-06,
      "loss": 0.0695,
      "step": 340
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 0.3425240218639374,
      "learning_rate": 9.230769230769232e-06,
      "loss": 0.1385,
      "step": 350
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 9.042716979980469,
      "learning_rate": 8.923076923076925e-06,
      "loss": 0.1137,
      "step": 360
    },
    {
      "epoch": 2.8461538461538463,
      "grad_norm": 0.414840966463089,
      "learning_rate": 8.615384615384617e-06,
      "loss": 0.1497,
      "step": 370
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 1.484980583190918,
      "learning_rate": 8.307692307692309e-06,
      "loss": 0.1326,
      "step": 380
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.287628710269928,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.1385,
      "step": 390
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.09617961943149567,
      "eval_runtime": 0.5589,
      "eval_samples_per_second": 237.964,
      "eval_steps_per_second": 30.416,
      "step": 390
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 2.3397181034088135,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.1411,
      "step": 400
    },
    {
      "epoch": 3.1538461538461537,
      "grad_norm": 5.8266520500183105,
      "learning_rate": 7.384615384615386e-06,
      "loss": 0.1549,
      "step": 410
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 9.101057052612305,
      "learning_rate": 7.076923076923078e-06,
      "loss": 0.1401,
      "step": 420
    },
    {
      "epoch": 3.3076923076923075,
      "grad_norm": 0.2339809238910675,
      "learning_rate": 6.76923076923077e-06,
      "loss": 0.0745,
      "step": 430
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 7.01936149597168,
      "learning_rate": 6.461538461538463e-06,
      "loss": 0.1161,
      "step": 440
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 12.716119766235352,
      "learning_rate": 6.153846153846155e-06,
      "loss": 0.0821,
      "step": 450
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 5.857155799865723,
      "learning_rate": 5.846153846153847e-06,
      "loss": 0.0956,
      "step": 460
    },
    {
      "epoch": 3.6153846153846154,
      "grad_norm": 7.989081859588623,
      "learning_rate": 5.538461538461539e-06,
      "loss": 0.0792,
      "step": 470
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.2197500616312027,
      "learning_rate": 5.230769230769232e-06,
      "loss": 0.0932,
      "step": 480
    },
    {
      "epoch": 3.769230769230769,
      "grad_norm": 0.21175715327262878,
      "learning_rate": 4.923076923076924e-06,
      "loss": 0.0733,
      "step": 490
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.9006862044334412,
      "learning_rate": 4.615384615384616e-06,
      "loss": 0.1399,
      "step": 500
    },
    {
      "epoch": 3.9230769230769234,
      "grad_norm": 0.20619560778141022,
      "learning_rate": 4.307692307692308e-06,
      "loss": 0.1307,
      "step": 510
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.3052387833595276,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0764,
      "step": 520
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9924812030075187,
      "eval_loss": 0.06696975976228714,
      "eval_runtime": 0.5431,
      "eval_samples_per_second": 244.875,
      "eval_steps_per_second": 31.3,
      "step": 520
    },
    {
      "epoch": 4.076923076923077,
      "grad_norm": 6.454896450042725,
      "learning_rate": 3.692307692307693e-06,
      "loss": 0.0969,
      "step": 530
    },
    {
      "epoch": 4.153846153846154,
      "grad_norm": 1.7816344499588013,
      "learning_rate": 3.384615384615385e-06,
      "loss": 0.1669,
      "step": 540
    },
    {
      "epoch": 4.230769230769231,
      "grad_norm": 0.3475962281227112,
      "learning_rate": 3.0769230769230774e-06,
      "loss": 0.1516,
      "step": 550
    },
    {
      "epoch": 4.3076923076923075,
      "grad_norm": 1.2676594257354736,
      "learning_rate": 2.7692307692307697e-06,
      "loss": 0.0501,
      "step": 560
    },
    {
      "epoch": 4.384615384615385,
      "grad_norm": 0.2670779526233673,
      "learning_rate": 2.461538461538462e-06,
      "loss": 0.1315,
      "step": 570
    },
    {
      "epoch": 4.461538461538462,
      "grad_norm": 0.26637333631515503,
      "learning_rate": 2.153846153846154e-06,
      "loss": 0.0812,
      "step": 580
    },
    {
      "epoch": 4.538461538461538,
      "grad_norm": 9.306379318237305,
      "learning_rate": 1.8461538461538465e-06,
      "loss": 0.0728,
      "step": 590
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 2.812556266784668,
      "learning_rate": 1.5384615384615387e-06,
      "loss": 0.085,
      "step": 600
    },
    {
      "epoch": 4.6923076923076925,
      "grad_norm": 5.42959451675415,
      "learning_rate": 1.230769230769231e-06,
      "loss": 0.093,
      "step": 610
    },
    {
      "epoch": 4.769230769230769,
      "grad_norm": 0.2531764805316925,
      "learning_rate": 9.230769230769232e-07,
      "loss": 0.085,
      "step": 620
    },
    {
      "epoch": 4.846153846153846,
      "grad_norm": 0.2155565619468689,
      "learning_rate": 6.153846153846155e-07,
      "loss": 0.0648,
      "step": 630
    },
    {
      "epoch": 4.923076923076923,
      "grad_norm": 0.23103830218315125,
      "learning_rate": 3.0769230769230774e-07,
      "loss": 0.0768,
      "step": 640
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.6090343594551086,
      "learning_rate": 0.0,
      "loss": 0.1154,
      "step": 650
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.08088764548301697,
      "eval_runtime": 0.5614,
      "eval_samples_per_second": 236.913,
      "eval_steps_per_second": 30.282,
      "step": 650
    },
    {
      "epoch": 5.0,
      "step": 650,
      "total_flos": 4.006371770595533e+17,
      "train_loss": 0.21900572785964378,
      "train_runtime": 48.7431,
      "train_samples_per_second": 106.066,
      "train_steps_per_second": 13.335
    }
  ],
  "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
}