{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.152165276122368,
  "eval_steps": 50,
  "global_step": 725,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 3e-05,
      "loss": 0.9927,
      "step": 10
    },
    {
      "epoch": 0.03,
      "learning_rate": 3e-05,
      "loss": 0.1794,
      "step": 20
    },
    {
      "epoch": 0.05,
      "learning_rate": 3e-05,
      "loss": 0.1202,
      "step": 30
    },
    {
      "epoch": 0.06,
      "learning_rate": 3e-05,
      "loss": 0.0955,
      "step": 40
    },
    {
      "epoch": 0.08,
      "learning_rate": 3e-05,
      "loss": 0.0886,
      "step": 50
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.5773584905660377,
      "eval_f1_macro": 0.32220178110380887,
      "eval_f1_micro": 0.5773584905660377,
      "eval_loss": 0.10824274271726608,
      "eval_precision_macro": 0.3987863781703509,
      "eval_precision_micro": 0.5773584905660377,
      "eval_recall_macro": 0.31235484191508905,
      "eval_recall_micro": 0.5773584905660377,
      "eval_runtime": 68.6809,
      "eval_samples_per_second": 15.434,
      "eval_steps_per_second": 3.858,
      "step": 50
    },
    {
      "epoch": 0.1,
      "learning_rate": 3e-05,
      "loss": 0.2261,
      "step": 60
    },
    {
      "epoch": 0.11,
      "learning_rate": 3e-05,
      "loss": 0.0728,
      "step": 70
    },
    {
      "epoch": 0.13,
      "learning_rate": 3e-05,
      "loss": 0.0659,
      "step": 80
    },
    {
      "epoch": 0.14,
      "learning_rate": 3e-05,
      "loss": 0.051,
      "step": 90
    },
    {
      "epoch": 0.16,
      "learning_rate": 3e-05,
      "loss": 0.0572,
      "step": 100
    },
    {
      "epoch": 0.16,
      "eval_accuracy": 0.5877358490566038,
      "eval_f1_macro": 0.379683051721915,
      "eval_f1_micro": 0.5877358490566038,
      "eval_loss": 0.08316469192504883,
      "eval_precision_macro": 0.4716361677310224,
      "eval_precision_micro": 0.5877358490566038,
      "eval_recall_macro": 0.3681154625887901,
      "eval_recall_micro": 0.5877358490566038,
      "eval_runtime": 68.5728,
      "eval_samples_per_second": 15.458,
      "eval_steps_per_second": 3.865,
      "step": 100
    },
    {
      "epoch": 0.17,
      "learning_rate": 3e-05,
      "loss": 0.1229,
      "step": 110
    },
    {
      "epoch": 0.19,
      "learning_rate": 3e-05,
      "loss": 0.0561,
      "step": 120
    },
    {
      "epoch": 0.21,
      "learning_rate": 3e-05,
      "loss": 0.0549,
      "step": 130
    },
    {
      "epoch": 0.22,
      "learning_rate": 3e-05,
      "loss": 0.0562,
      "step": 140
    },
    {
      "epoch": 0.24,
      "learning_rate": 3e-05,
      "loss": 0.0496,
      "step": 150
    },
    {
      "epoch": 0.24,
      "eval_accuracy": 0.7311320754716981,
      "eval_f1_macro": 0.5702888507555568,
      "eval_f1_micro": 0.7311320754716981,
      "eval_loss": 0.05246850848197937,
      "eval_precision_macro": 0.5911132598584014,
      "eval_precision_micro": 0.7311320754716981,
      "eval_recall_macro": 0.5746634992387409,
      "eval_recall_micro": 0.7311320754716981,
      "eval_runtime": 68.5467,
      "eval_samples_per_second": 15.464,
      "eval_steps_per_second": 3.866,
      "step": 150
    },
    {
      "epoch": 0.25,
      "learning_rate": 3e-05,
      "loss": 0.067,
      "step": 160
    },
    {
      "epoch": 0.27,
      "learning_rate": 3e-05,
      "loss": 0.0551,
      "step": 170
    },
    {
      "epoch": 0.29,
      "learning_rate": 3e-05,
      "loss": 0.0426,
      "step": 180
    },
    {
      "epoch": 0.3,
      "learning_rate": 3e-05,
      "loss": 0.0475,
      "step": 190
    },
    {
      "epoch": 0.32,
      "learning_rate": 3e-05,
      "loss": 0.0541,
      "step": 200
    },
    {
      "epoch": 0.32,
      "eval_accuracy": 0.7566037735849057,
      "eval_f1_macro": 0.5584352420606399,
      "eval_f1_micro": 0.7566037735849057,
      "eval_loss": 0.04639677330851555,
      "eval_precision_macro": 0.6151030653662273,
      "eval_precision_micro": 0.7566037735849057,
      "eval_recall_macro": 0.5606200967693734,
      "eval_recall_micro": 0.7566037735849057,
      "eval_runtime": 68.5714,
      "eval_samples_per_second": 15.458,
      "eval_steps_per_second": 3.865,
      "step": 200
    },
    {
      "epoch": 0.33,
      "learning_rate": 3e-05,
      "loss": 0.052,
      "step": 210
    },
    {
      "epoch": 0.35,
      "learning_rate": 3e-05,
      "loss": 0.0435,
      "step": 220
    },
    {
      "epoch": 0.37,
      "learning_rate": 3e-05,
      "loss": 0.0407,
      "step": 230
    },
    {
      "epoch": 0.38,
      "learning_rate": 3e-05,
      "loss": 0.0475,
      "step": 240
    },
    {
      "epoch": 0.4,
      "learning_rate": 3e-05,
      "loss": 0.0481,
      "step": 250
    },
    {
      "epoch": 0.4,
      "eval_accuracy": 0.7811320754716982,
      "eval_f1_macro": 0.6368753336945477,
      "eval_f1_micro": 0.7811320754716982,
      "eval_loss": 0.04328591376543045,
      "eval_precision_macro": 0.6636054486047108,
      "eval_precision_micro": 0.7811320754716982,
      "eval_recall_macro": 0.6513962288983521,
      "eval_recall_micro": 0.7811320754716982,
      "eval_runtime": 68.5331,
      "eval_samples_per_second": 15.467,
      "eval_steps_per_second": 3.867,
      "step": 250
    },
    {
      "epoch": 0.41,
      "learning_rate": 3e-05,
      "loss": 0.0533,
      "step": 260
    },
    {
      "epoch": 0.43,
      "learning_rate": 3e-05,
      "loss": 0.0433,
      "step": 270
    },
    {
      "epoch": 0.44,
      "learning_rate": 3e-05,
      "loss": 0.0432,
      "step": 280
    },
    {
      "epoch": 0.46,
      "learning_rate": 3e-05,
      "loss": 0.0466,
      "step": 290
    },
    {
      "epoch": 0.48,
      "learning_rate": 3e-05,
      "loss": 0.053,
      "step": 300
    },
    {
      "epoch": 0.48,
      "eval_accuracy": 0.7632075471698113,
      "eval_f1_macro": 0.6337788769448006,
      "eval_f1_micro": 0.7632075471698113,
      "eval_loss": 0.0451766662299633,
      "eval_precision_macro": 0.6935549047637881,
      "eval_precision_micro": 0.7632075471698113,
      "eval_recall_macro": 0.6461210681607904,
      "eval_recall_micro": 0.7632075471698113,
      "eval_runtime": 68.6668,
      "eval_samples_per_second": 15.437,
      "eval_steps_per_second": 3.859,
      "step": 300
    },
    {
      "epoch": 0.49,
      "learning_rate": 3e-05,
      "loss": 0.0455,
      "step": 310
    },
    {
      "epoch": 0.51,
      "learning_rate": 3e-05,
      "loss": 0.049,
      "step": 320
    },
    {
      "epoch": 0.52,
      "learning_rate": 3e-05,
      "loss": 0.0426,
      "step": 330
    },
    {
      "epoch": 0.54,
      "learning_rate": 3e-05,
      "loss": 0.0396,
      "step": 340
    },
    {
      "epoch": 0.56,
      "learning_rate": 3e-05,
      "loss": 0.0401,
      "step": 350
    },
    {
      "epoch": 0.56,
      "eval_accuracy": 0.7943396226415095,
      "eval_f1_macro": 0.6696815276160976,
      "eval_f1_micro": 0.7943396226415095,
      "eval_loss": 0.039866555482149124,
      "eval_precision_macro": 0.7380690805686413,
      "eval_precision_micro": 0.7943396226415095,
      "eval_recall_macro": 0.660366139065656,
      "eval_recall_micro": 0.7943396226415095,
      "eval_runtime": 68.673,
      "eval_samples_per_second": 15.435,
      "eval_steps_per_second": 3.859,
      "step": 350
    },
    {
      "epoch": 0.57,
      "learning_rate": 3e-05,
      "loss": 0.0482,
      "step": 360
    },
    {
      "epoch": 0.59,
      "learning_rate": 3e-05,
      "loss": 0.0373,
      "step": 370
    },
    {
      "epoch": 0.6,
      "learning_rate": 3e-05,
      "loss": 0.036,
      "step": 380
    },
    {
      "epoch": 0.62,
      "learning_rate": 3e-05,
      "loss": 0.0428,
      "step": 390
    },
    {
      "epoch": 0.64,
      "learning_rate": 3e-05,
      "loss": 0.0509,
      "step": 400
    },
    {
      "epoch": 0.64,
      "eval_accuracy": 0.8009433962264151,
      "eval_f1_macro": 0.6501313860427956,
      "eval_f1_micro": 0.8009433962264151,
      "eval_loss": 0.03930637985467911,
      "eval_precision_macro": 0.6546476325081232,
      "eval_precision_micro": 0.8009433962264151,
      "eval_recall_macro": 0.6611935636788673,
      "eval_recall_micro": 0.8009433962264151,
      "eval_runtime": 68.557,
      "eval_samples_per_second": 15.462,
      "eval_steps_per_second": 3.865,
      "step": 400
    },
    {
      "epoch": 0.65,
      "learning_rate": 3e-05,
      "loss": 0.0375,
      "step": 410
    },
    {
      "epoch": 0.67,
      "learning_rate": 3e-05,
      "loss": 0.041,
      "step": 420
    },
    {
      "epoch": 0.68,
      "learning_rate": 3e-05,
      "loss": 0.0416,
      "step": 430
    },
    {
      "epoch": 0.7,
      "learning_rate": 3e-05,
      "loss": 0.0396,
      "step": 440
    },
    {
      "epoch": 0.72,
      "learning_rate": 3e-05,
      "loss": 0.0474,
      "step": 450
    },
    {
      "epoch": 0.72,
      "eval_accuracy": 0.8018867924528302,
      "eval_f1_macro": 0.6864569711826704,
      "eval_f1_micro": 0.8018867924528302,
      "eval_loss": 0.04012966528534889,
      "eval_precision_macro": 0.7255429634365795,
      "eval_precision_micro": 0.8018867924528302,
      "eval_recall_macro": 0.6926779368041328,
      "eval_recall_micro": 0.8018867924528302,
      "eval_runtime": 68.6193,
      "eval_samples_per_second": 15.448,
      "eval_steps_per_second": 3.862,
      "step": 450
    },
    {
      "epoch": 0.73,
      "learning_rate": 3e-05,
      "loss": 0.0434,
      "step": 460
    },
    {
      "epoch": 0.75,
      "learning_rate": 3e-05,
      "loss": 0.0358,
      "step": 470
    },
    {
      "epoch": 0.76,
      "learning_rate": 3e-05,
      "loss": 0.0416,
      "step": 480
    },
    {
      "epoch": 0.78,
      "learning_rate": 3e-05,
      "loss": 0.0334,
      "step": 490
    },
    {
      "epoch": 0.79,
      "learning_rate": 3e-05,
      "loss": 0.045,
      "step": 500
    },
    {
      "epoch": 0.79,
      "eval_accuracy": 0.8009433962264151,
      "eval_f1_macro": 0.6977412107603574,
      "eval_f1_micro": 0.8009433962264151,
      "eval_loss": 0.0379195362329483,
      "eval_precision_macro": 0.7146704097806501,
      "eval_precision_micro": 0.8009433962264151,
      "eval_recall_macro": 0.710805016133364,
      "eval_recall_micro": 0.8009433962264151,
      "eval_runtime": 68.6344,
      "eval_samples_per_second": 15.444,
      "eval_steps_per_second": 3.861,
      "step": 500
    },
    {
      "epoch": 0.81,
      "learning_rate": 3e-05,
      "loss": 0.0425,
      "step": 510
    },
    {
      "epoch": 0.83,
      "learning_rate": 3e-05,
      "loss": 0.036,
      "step": 520
    },
    {
      "epoch": 0.84,
      "learning_rate": 3e-05,
      "loss": 0.0444,
      "step": 530
    },
    {
      "epoch": 0.86,
      "learning_rate": 3e-05,
      "loss": 0.0394,
      "step": 540
    },
    {
      "epoch": 0.87,
      "learning_rate": 3e-05,
      "loss": 0.0335,
      "step": 550
    },
    {
      "epoch": 0.87,
      "eval_accuracy": 0.8150943396226416,
      "eval_f1_macro": 0.7134710025440245,
      "eval_f1_micro": 0.8150943396226416,
      "eval_loss": 0.03691105917096138,
      "eval_precision_macro": 0.7046165923538829,
      "eval_precision_micro": 0.8150943396226416,
      "eval_recall_macro": 0.7335435173781253,
      "eval_recall_micro": 0.8150943396226416,
      "eval_runtime": 68.5596,
      "eval_samples_per_second": 15.461,
      "eval_steps_per_second": 3.865,
      "step": 550
    },
    {
      "epoch": 0.89,
      "learning_rate": 3e-05,
      "loss": 0.0421,
      "step": 560
    },
    {
      "epoch": 0.91,
      "learning_rate": 3e-05,
      "loss": 0.0407,
      "step": 570
    },
    {
      "epoch": 0.92,
      "learning_rate": 3e-05,
      "loss": 0.0429,
      "step": 580
    },
    {
      "epoch": 0.94,
      "learning_rate": 3e-05,
      "loss": 0.0378,
      "step": 590
    },
    {
      "epoch": 0.95,
      "learning_rate": 3e-05,
      "loss": 0.0429,
      "step": 600
    },
    {
      "epoch": 0.95,
      "eval_accuracy": 0.7962264150943397,
      "eval_f1_macro": 0.687832173620461,
      "eval_f1_micro": 0.7962264150943396,
      "eval_loss": 0.03668661788105965,
      "eval_precision_macro": 0.7081030422724828,
      "eval_precision_micro": 0.7962264150943397,
      "eval_recall_macro": 0.6958905634881637,
      "eval_recall_micro": 0.7962264150943397,
      "eval_runtime": 68.6225,
      "eval_samples_per_second": 15.447,
      "eval_steps_per_second": 3.862,
      "step": 600
    },
    {
      "epoch": 0.97,
      "learning_rate": 3e-05,
      "loss": 0.0394,
      "step": 610
    },
    {
      "epoch": 0.99,
      "learning_rate": 3e-05,
      "loss": 0.0363,
      "step": 620
    },
    {
      "epoch": 1.0,
      "learning_rate": 3e-05,
      "loss": 0.0428,
      "step": 630
    },
    {
      "epoch": 1.02,
      "learning_rate": 3e-05,
      "loss": 0.0265,
      "step": 640
    },
    {
      "epoch": 1.03,
      "learning_rate": 3e-05,
      "loss": 0.0253,
      "step": 650
    },
    {
      "epoch": 1.03,
      "eval_accuracy": 0.8254716981132075,
      "eval_f1_macro": 0.7098363543260356,
      "eval_f1_micro": 0.8254716981132075,
      "eval_loss": 0.03421418368816376,
      "eval_precision_macro": 0.7370277877974788,
      "eval_precision_micro": 0.8254716981132075,
      "eval_recall_macro": 0.6974501533785277,
      "eval_recall_micro": 0.8254716981132075,
      "eval_runtime": 68.5452,
      "eval_samples_per_second": 15.464,
      "eval_steps_per_second": 3.866,
      "step": 650
    },
    {
      "epoch": 1.05,
      "learning_rate": 3e-05,
      "loss": 0.0243,
      "step": 660
    },
    {
      "epoch": 1.06,
      "learning_rate": 3e-05,
      "loss": 0.0313,
      "step": 670
    },
    {
      "epoch": 1.08,
      "learning_rate": 3e-05,
      "loss": 0.0285,
      "step": 680
    },
    {
      "epoch": 1.1,
      "learning_rate": 3e-05,
      "loss": 0.0262,
      "step": 690
    },
    {
      "epoch": 1.11,
      "learning_rate": 3e-05,
      "loss": 0.0311,
      "step": 700
    },
    {
      "epoch": 1.11,
      "eval_accuracy": 0.8047169811320755,
      "eval_f1_macro": 0.6661175684975367,
      "eval_f1_micro": 0.8047169811320755,
      "eval_loss": 0.0357016883790493,
      "eval_precision_macro": 0.6994736595477448,
      "eval_precision_micro": 0.8047169811320755,
      "eval_recall_macro": 0.660920267471505,
      "eval_recall_micro": 0.8047169811320755,
      "eval_runtime": 68.6332,
      "eval_samples_per_second": 15.444,
      "eval_steps_per_second": 3.861,
      "step": 700
    },
    {
      "epoch": 1.13,
      "learning_rate": 3e-05,
      "loss": 0.0311,
      "step": 710
    },
    {
      "epoch": 1.14,
      "learning_rate": 3e-05,
      "loss": 0.0298,
      "step": 720
    },
    {
      "epoch": 1.15,
      "step": 725,
      "total_flos": 3.949643257934285e+17,
      "train_loss": 0.06415341473858932,
      "train_runtime": 4786.626,
      "train_samples_per_second": 4.847,
      "train_steps_per_second": 0.151
    }
  ],
  "logging_steps": 10,
  "max_steps": 725,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 250,
  "total_flos": 3.949643257934285e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}