{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.0,
  "global_step": 26299,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.27,
      "learning_rate": 1.923951481044907e-05,
      "loss": 0.0205,
      "step": 1000
    },
    {
      "epoch": 0.27,
      "eval_EG_f1": 0.9960332310455804,
      "eval_ET_f1": 0.9988771614641814,
      "eval_TE_f1": 0.9952833720146739,
      "eval_loss": 0.0021067976485937834,
      "eval_overall_accuracy": 0.9995180550845189,
      "eval_overall_f1": 0.9967312922623949,
      "eval_overall_precision": 0.9966566866267466,
      "eval_overall_recall": 0.9968059090682237,
      "eval_runtime": 16.2281,
      "eval_samples_per_second": 411.569,
      "eval_steps_per_second": 25.758,
      "step": 1000
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.8479029620898135e-05,
      "loss": 0.0027,
      "step": 2000
    },
    {
      "epoch": 0.53,
      "eval_EG_f1": 0.9955841628620612,
      "eval_ET_f1": 0.9992513849378649,
      "eval_TE_f1": 0.995433789954338,
      "eval_loss": 0.0024818568490445614,
      "eval_overall_accuracy": 0.9993931064027275,
      "eval_overall_f1": 0.9967563251659264,
      "eval_overall_precision": 0.9966568534504266,
      "eval_overall_recall": 0.9968558167390328,
      "eval_runtime": 16.4446,
      "eval_samples_per_second": 406.152,
      "eval_steps_per_second": 25.419,
      "step": 2000
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.77185444313472e-05,
      "loss": 0.0018,
      "step": 3000
    },
    {
      "epoch": 0.8,
      "eval_EG_f1": 0.9987274496594056,
      "eval_ET_f1": 0.999700553975146,
      "eval_TE_f1": 0.9982783142450782,
      "eval_loss": 0.0007141608512029052,
      "eval_overall_accuracy": 0.9998393516948396,
      "eval_overall_f1": 0.9989020860365306,
      "eval_overall_precision": 0.9988522381356355,
      "eval_overall_recall": 0.9989519389130109,
      "eval_runtime": 16.3629,
      "eval_samples_per_second": 408.181,
      "eval_steps_per_second": 25.546,
      "step": 3000
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.695805924179627e-05,
      "loss": 0.0015,
      "step": 4000
    },
    {
      "epoch": 1.06,
      "eval_EG_f1": 0.9991765850737331,
      "eval_ET_f1": 0.9992513849378649,
      "eval_TE_f1": 0.9983530468633028,
      "eval_loss": 0.0010974227916449308,
      "eval_overall_accuracy": 0.9997858022597862,
      "eval_overall_f1": 0.9989270118527761,
      "eval_overall_precision": 0.9989020860365306,
      "eval_overall_recall": 0.9989519389130109,
      "eval_runtime": 16.7722,
      "eval_samples_per_second": 398.219,
      "eval_steps_per_second": 24.922,
      "step": 4000
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.6197574052245332e-05,
      "loss": 0.0005,
      "step": 5000
    },
    {
      "epoch": 1.33,
      "eval_EG_f1": 0.9982030548068285,
      "eval_ET_f1": 0.999850276987573,
      "eval_TE_f1": 0.9979038778260219,
      "eval_loss": 0.0007138435612432659,
      "eval_overall_accuracy": 0.9998125769773129,
      "eval_overall_f1": 0.998652425633859,
      "eval_overall_precision": 0.998702271025705,
      "eval_overall_recall": 0.9986025852173479,
      "eval_runtime": 16.5358,
      "eval_samples_per_second": 403.912,
      "eval_steps_per_second": 25.279,
      "step": 5000
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.54370888626944e-05,
      "loss": 0.0008,
      "step": 6000
    },
    {
      "epoch": 1.6,
      "eval_EG_f1": 0.9983530468633028,
      "eval_ET_f1": 0.999550830962719,
      "eval_TE_f1": 0.9979038778260219,
      "eval_loss": 0.0009068112121894956,
      "eval_overall_accuracy": 0.9997947271656284,
      "eval_overall_f1": 0.9986025852173479,
      "eval_overall_precision": 0.9986025852173479,
      "eval_overall_recall": 0.9986025852173479,
      "eval_runtime": 16.3597,
      "eval_samples_per_second": 408.259,
      "eval_steps_per_second": 25.551,
      "step": 6000
    },
    {
      "epoch": 1.86,
      "learning_rate": 1.4676603673143466e-05,
      "loss": 0.0009,
      "step": 7000
    },
    {
      "epoch": 1.86,
      "eval_EG_f1": 0.9989519389130109,
      "eval_ET_f1": 0.999850276987573,
      "eval_TE_f1": 0.9988022159005839,
      "eval_loss": 0.0010577181819826365,
      "eval_overall_accuracy": 0.9998482766006819,
      "eval_overall_f1": 0.999201477267056,
      "eval_overall_precision": 0.999201477267056,
      "eval_overall_recall": 0.999201477267056,
      "eval_runtime": 16.3447,
      "eval_samples_per_second": 408.634,
      "eval_steps_per_second": 25.574,
      "step": 7000
    },
    {
      "epoch": 2.13,
      "learning_rate": 1.3916118483592534e-05,
      "loss": 0.0005,
      "step": 8000
    },
    {
      "epoch": 2.13,
      "eval_EG_f1": 0.9991765850737331,
      "eval_ET_f1": 0.999850276987573,
      "eval_TE_f1": 0.9991765850737331,
      "eval_loss": 0.0005780701176263392,
      "eval_overall_accuracy": 0.9998839762240508,
      "eval_overall_f1": 0.9994011378381076,
      "eval_overall_precision": 0.9993512650331853,
      "eval_overall_recall": 0.999451015621101,
      "eval_runtime": 16.6712,
      "eval_samples_per_second": 400.631,
      "eval_steps_per_second": 25.073,
      "step": 8000
    },
    {
      "epoch": 2.4,
      "learning_rate": 1.31556332940416e-05,
      "loss": 0.0002,
      "step": 9000
    },
    {
      "epoch": 2.4,
      "eval_EG_f1": 0.9988771614641814,
      "eval_ET_f1": 0.999850276987573,
      "eval_TE_f1": 0.9985777378546299,
      "eval_loss": 0.0005696099251508713,
      "eval_overall_accuracy": 0.999892901129893,
      "eval_overall_f1": 0.9991017067571614,
      "eval_overall_precision": 0.9990518488946554,
      "eval_overall_recall": 0.999151569596247,
      "eval_runtime": 16.4448,
      "eval_samples_per_second": 406.147,
      "eval_steps_per_second": 25.418,
      "step": 9000
    },
    {
      "epoch": 2.66,
      "learning_rate": 1.2395148104490666e-05,
      "loss": 0.0005,
      "step": 10000
    },
    {
      "epoch": 2.66,
      "eval_EG_f1": 0.9989519389130109,
      "eval_ET_f1": 0.999850276987573,
      "eval_TE_f1": 0.9988771614641814,
      "eval_loss": 0.0004689108463935554,
      "eval_overall_accuracy": 0.999892901129893,
      "eval_overall_f1": 0.9992264504054896,
      "eval_overall_precision": 0.9992015171174768,
      "eval_overall_recall": 0.9992513849378649,
      "eval_runtime": 16.3326,
      "eval_samples_per_second": 408.938,
      "eval_steps_per_second": 25.593,
      "step": 10000
    },
    {
      "epoch": 2.93,
      "learning_rate": 1.1634662914939731e-05,
      "loss": 0.0002,
      "step": 11000
    },
    {
      "epoch": 2.93,
      "eval_EG_f1": 0.9988771614641814,
      "eval_ET_f1": 1.0,
      "eval_TE_f1": 0.9990268732689572,
      "eval_loss": 0.0005850127199664712,
      "eval_overall_accuracy": 0.9999286007532621,
      "eval_overall_f1": 0.9993013274777922,
      "eval_overall_precision": 0.9992514596536753,
      "eval_overall_recall": 0.999351200279483,
      "eval_runtime": 16.4437,
      "eval_samples_per_second": 406.175,
      "eval_steps_per_second": 25.42,
      "step": 11000
    },
    {
      "epoch": 3.19,
      "learning_rate": 1.08741777253888e-05,
      "loss": 0.0008,
      "step": 12000
    },
    {
      "epoch": 3.19,
      "eval_EG_f1": 0.9992513849378649,
      "eval_ET_f1": 1.0,
      "eval_TE_f1": 0.9992513849378649,
      "eval_loss": 0.0004198316019028425,
      "eval_overall_accuracy": 0.9999464505649466,
      "eval_overall_f1": 0.99950092329191,
      "eval_overall_precision": 0.99950092329191,
      "eval_overall_recall": 0.99950092329191,
      "eval_runtime": 16.4892,
      "eval_samples_per_second": 405.052,
      "eval_steps_per_second": 25.35,
      "step": 12000
    },
    {
      "epoch": 3.46,
      "learning_rate": 1.0113692535837865e-05,
      "loss": 0.0002,
      "step": 13000
    },
    {
      "epoch": 3.46,
      "eval_EG_f1": 0.9991016619254379,
      "eval_ET_f1": 0.999700553975146,
      "eval_TE_f1": 0.9989519389130109,
      "eval_loss": 0.0011158857960253954,
      "eval_overall_accuracy": 0.9998482766006819,
      "eval_overall_f1": 0.9992513849378649,
      "eval_overall_precision": 0.9992513849378649,
      "eval_overall_recall": 0.9992513849378649,
      "eval_runtime": 16.7537,
      "eval_samples_per_second": 398.659,
      "eval_steps_per_second": 24.95,
      "step": 13000
    },
    {
      "epoch": 3.73,
      "learning_rate": 9.353207346286931e-06,
      "loss": 0.0003,
      "step": 14000
    },
    {
      "epoch": 3.73,
      "eval_EG_f1": 0.9988022159005839,
      "eval_ET_f1": 0.999850276987573,
      "eval_TE_f1": 0.9986524928881569,
      "eval_loss": 0.0010174426715821028,
      "eval_overall_accuracy": 0.9998393516948396,
      "eval_overall_f1": 0.9991016619254379,
      "eval_overall_precision": 0.9991016619254379,
      "eval_overall_recall": 0.9991016619254379,
      "eval_runtime": 16.5225,
      "eval_samples_per_second": 404.237,
      "eval_steps_per_second": 25.299,
      "step": 14000
    },
    {
      "epoch": 3.99,
      "learning_rate": 8.592722156735998e-06,
      "loss": 0.0,
      "step": 15000
    },
    {
      "epoch": 3.99,
      "eval_EG_f1": 0.9992513849378649,
      "eval_ET_f1": 1.0,
      "eval_TE_f1": 0.9992513849378649,
      "eval_loss": 0.001382750691846013,
      "eval_overall_accuracy": 0.9999018260357353,
      "eval_overall_f1": 0.99950092329191,
      "eval_overall_precision": 0.99950092329191,
      "eval_overall_recall": 0.99950092329191,
      "eval_runtime": 16.3741,
      "eval_samples_per_second": 407.899,
      "eval_steps_per_second": 25.528,
      "step": 15000
    },
    {
      "epoch": 4.26,
      "learning_rate": 7.832236967185065e-06,
      "loss": 0.0002,
      "step": 16000
    },
    {
      "epoch": 4.26,
      "eval_EG_f1": 0.9990268732689572,
      "eval_ET_f1": 0.9997754322928364,
      "eval_TE_f1": 0.9991016619254379,
      "eval_loss": 0.0008425627020187676,
      "eval_overall_accuracy": 0.9998839762240508,
      "eval_overall_f1": 0.9993013274777922,
      "eval_overall_precision": 0.9992514596536753,
      "eval_overall_recall": 0.999351200279483,
      "eval_runtime": 16.4088,
      "eval_samples_per_second": 407.037,
      "eval_steps_per_second": 25.474,
      "step": 16000
    },
    {
      "epoch": 4.52,
      "learning_rate": 7.071751777634132e-06,
      "loss": 0.0001,
      "step": 17000
    },
    {
      "epoch": 4.52,
      "eval_EG_f1": 0.9994011079502919,
      "eval_ET_f1": 0.999700553975146,
      "eval_TE_f1": 0.9992513849378649,
      "eval_loss": 0.000884207256603986,
      "eval_overall_accuracy": 0.9999196758474198,
      "eval_overall_f1": 0.999451015621101,
      "eval_overall_precision": 0.999451015621101,
      "eval_overall_recall": 0.999451015621101,
      "eval_runtime": 16.6905,
      "eval_samples_per_second": 400.169,
      "eval_steps_per_second": 25.044,
      "step": 17000
    },
    {
      "epoch": 4.79,
      "learning_rate": 6.311266588083198e-06,
      "loss": 0.0005,
      "step": 18000
    },
    {
      "epoch": 4.79,
      "eval_EG_f1": 0.9992513849378649,
      "eval_ET_f1": 1.0,
      "eval_TE_f1": 0.9992513849378649,
      "eval_loss": 0.0005574871320277452,
      "eval_overall_accuracy": 0.9999375256591043,
      "eval_overall_f1": 0.99950092329191,
      "eval_overall_precision": 0.99950092329191,
      "eval_overall_recall": 0.99950092329191,
      "eval_runtime": 16.5015,
      "eval_samples_per_second": 404.752,
      "eval_steps_per_second": 25.331,
      "step": 18000
    },
    {
      "epoch": 5.06,
      "learning_rate": 5.5507813985322644e-06,
      "loss": 0.0,
      "step": 19000
    },
    {
      "epoch": 5.06,
      "eval_EG_f1": 0.9989519389130109,
      "eval_ET_f1": 1.0,
      "eval_TE_f1": 0.9989519389130109,
      "eval_loss": 0.0005674651474691927,
      "eval_overall_accuracy": 0.9999018260357353,
      "eval_overall_f1": 0.999301292608674,
      "eval_overall_precision": 0.999301292608674,
      "eval_overall_recall": 0.999301292608674,
      "eval_runtime": 16.4647,
      "eval_samples_per_second": 405.657,
      "eval_steps_per_second": 25.388,
      "step": 19000
    },
    {
      "epoch": 5.32,
      "learning_rate": 4.79029620898133e-06,
      "loss": 0.0001,
      "step": 20000
    },
    {
      "epoch": 5.32,
      "eval_EG_f1": 0.9992513849378649,
      "eval_ET_f1": 1.0,
      "eval_TE_f1": 0.9992513849378649,
      "eval_loss": 0.0006091786199249327,
      "eval_overall_accuracy": 0.9999375256591043,
      "eval_overall_f1": 0.99950092329191,
      "eval_overall_precision": 0.99950092329191,
      "eval_overall_recall": 0.99950092329191,
      "eval_runtime": 16.4349,
      "eval_samples_per_second": 406.391,
      "eval_steps_per_second": 25.434,
      "step": 20000
    },
    {
      "epoch": 5.59,
      "learning_rate": 4.029811019430397e-06,
      "loss": 0.0,
      "step": 21000
    },
    {
      "epoch": 5.59,
      "eval_EG_f1": 0.9994011079502919,
      "eval_ET_f1": 1.0,
      "eval_TE_f1": 0.9994011079502919,
      "eval_loss": 0.0006266526179388165,
      "eval_overall_accuracy": 0.9999553754707888,
      "eval_overall_f1": 0.9996007386335279,
      "eval_overall_precision": 0.9996007386335279,
      "eval_overall_recall": 0.9996007386335279,
      "eval_runtime": 16.6877,
      "eval_samples_per_second": 400.236,
      "eval_steps_per_second": 25.048,
      "step": 21000
    },
    {
      "epoch": 5.86,
      "learning_rate": 3.2693258298794635e-06,
      "loss": 0.0,
      "step": 22000
    },
    {
      "epoch": 5.86,
      "eval_EG_f1": 0.9992513849378649,
      "eval_ET_f1": 1.0,
      "eval_TE_f1": 0.9992513849378649,
      "eval_loss": 0.0006635423633269966,
      "eval_overall_accuracy": 0.9999375256591043,
      "eval_overall_f1": 0.99950092329191,
      "eval_overall_precision": 0.99950092329191,
      "eval_overall_recall": 0.99950092329191,
      "eval_runtime": 16.716,
      "eval_samples_per_second": 399.557,
      "eval_steps_per_second": 25.006,
      "step": 22000
    },
    {
      "epoch": 6.12,
      "learning_rate": 2.50884064032853e-06,
      "loss": 0.0,
      "step": 23000
    },
    {
      "epoch": 6.12,
      "eval_EG_f1": 0.9994011079502919,
      "eval_ET_f1": 1.0,
      "eval_TE_f1": 0.9994011079502919,
      "eval_loss": 0.0006512191030196846,
      "eval_overall_accuracy": 0.9999464505649466,
      "eval_overall_f1": 0.9996007386335279,
      "eval_overall_precision": 0.9996007386335279,
      "eval_overall_recall": 0.9996007386335279,
      "eval_runtime": 16.3818,
      "eval_samples_per_second": 407.709,
      "eval_steps_per_second": 25.516,
      "step": 23000
    },
    {
      "epoch": 6.39,
      "learning_rate": 1.7483554507775962e-06,
      "loss": 0.0,
      "step": 24000
    },
    {
      "epoch": 6.39,
      "eval_EG_f1": 0.9994011079502919,
      "eval_ET_f1": 1.0,
      "eval_TE_f1": 0.9994011079502919,
      "eval_loss": 0.000729710329324007,
      "eval_overall_accuracy": 0.9999464505649466,
      "eval_overall_f1": 0.9996007386335279,
      "eval_overall_precision": 0.9996007386335279,
      "eval_overall_recall": 0.9996007386335279,
      "eval_runtime": 16.5641,
      "eval_samples_per_second": 403.222,
      "eval_steps_per_second": 25.235,
      "step": 24000
    },
    {
      "epoch": 6.65,
      "learning_rate": 9.878702612266627e-07,
      "loss": 0.0001,
      "step": 25000
    },
    {
      "epoch": 6.65,
      "eval_EG_f1": 0.9994011079502919,
      "eval_ET_f1": 1.0,
      "eval_TE_f1": 0.9994011079502919,
      "eval_loss": 0.0005905419238843024,
      "eval_overall_accuracy": 0.9999464505649466,
      "eval_overall_f1": 0.9996007386335279,
      "eval_overall_precision": 0.9996007386335279,
      "eval_overall_recall": 0.9996007386335279,
      "eval_runtime": 16.5058,
      "eval_samples_per_second": 404.647,
      "eval_steps_per_second": 25.325,
      "step": 25000
    },
    {
      "epoch": 6.92,
      "learning_rate": 2.2738507167572915e-07,
      "loss": 0.0001,
      "step": 26000
    },
    {
      "epoch": 6.92,
      "eval_EG_f1": 0.9994011079502919,
      "eval_ET_f1": 1.0,
      "eval_TE_f1": 0.9994011079502919,
      "eval_loss": 0.0006466144695878029,
      "eval_overall_accuracy": 0.9999464505649466,
      "eval_overall_f1": 0.9996007386335279,
      "eval_overall_precision": 0.9996007386335279,
      "eval_overall_recall": 0.9996007386335279,
      "eval_runtime": 16.8702,
      "eval_samples_per_second": 395.906,
      "eval_steps_per_second": 24.777,
      "step": 26000
    }
  ],
  "max_steps": 26299,
  "num_train_epochs": 7,
  "total_flos": 6405072201643752.0,
  "trial_name": null,
  "trial_params": null
}