{
  "best_metric": 0.8198928704505658,
  "best_model_checkpoint": "/root/kyivnotkiev/data/cl/model/benchmark_runs/deberta-v3-large_lr1e-05_seed456_weighted/checkpoint-1840",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1840,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02717391304347826,
      "grad_norm": 6.552072048187256,
      "learning_rate": 9.057971014492754e-07,
      "loss": 2.4236,
      "step": 50
    },
    {
      "epoch": 0.05434782608695652,
      "grad_norm": 5.586729049682617,
      "learning_rate": 1.8115942028985508e-06,
      "loss": 2.4231,
      "step": 100
    },
    {
      "epoch": 0.08152173913043478,
      "grad_norm": 3.997156858444214,
      "learning_rate": 2.6992753623188405e-06,
      "loss": 2.402,
      "step": 150
    },
    {
      "epoch": 0.10869565217391304,
      "grad_norm": 6.317577362060547,
      "learning_rate": 3.6050724637681163e-06,
      "loss": 2.3855,
      "step": 200
    },
    {
      "epoch": 0.1358695652173913,
      "grad_norm": 9.232672691345215,
      "learning_rate": 4.510869565217392e-06,
      "loss": 2.2186,
      "step": 250
    },
    {
      "epoch": 0.16304347826086957,
      "grad_norm": 9.297445297241211,
      "learning_rate": 5.416666666666667e-06,
      "loss": 1.9197,
      "step": 300
    },
    {
      "epoch": 0.19021739130434784,
      "grad_norm": 9.51098346710205,
      "learning_rate": 6.322463768115943e-06,
      "loss": 1.6219,
      "step": 350
    },
    {
      "epoch": 0.21739130434782608,
      "grad_norm": 8.97563648223877,
      "learning_rate": 7.228260869565218e-06,
      "loss": 1.1933,
      "step": 400
    },
    {
      "epoch": 0.24456521739130435,
      "grad_norm": 17.16935157775879,
      "learning_rate": 8.115942028985508e-06,
      "loss": 0.9993,
      "step": 450
    },
    {
      "epoch": 0.2717391304347826,
      "grad_norm": 19.718891143798828,
      "learning_rate": 9.021739130434784e-06,
      "loss": 0.8655,
      "step": 500
    },
    {
      "epoch": 0.29891304347826086,
      "grad_norm": 11.259733200073242,
      "learning_rate": 9.927536231884058e-06,
      "loss": 0.8272,
      "step": 550
    },
    {
      "epoch": 0.32608695652173914,
      "grad_norm": 24.362035751342773,
      "learning_rate": 9.909420289855073e-06,
      "loss": 0.6588,
      "step": 600
    },
    {
      "epoch": 0.3532608695652174,
      "grad_norm": 14.829962730407715,
      "learning_rate": 9.808776167471822e-06,
      "loss": 0.7243,
      "step": 650
    },
    {
      "epoch": 0.3804347826086957,
      "grad_norm": 21.006099700927734,
      "learning_rate": 9.710144927536233e-06,
      "loss": 0.7559,
      "step": 700
    },
    {
      "epoch": 0.4076086956521739,
      "grad_norm": 4.0826263427734375,
      "learning_rate": 9.60950080515298e-06,
      "loss": 0.7237,
      "step": 750
    },
    {
      "epoch": 0.43478260869565216,
      "grad_norm": 16.393157958984375,
      "learning_rate": 9.508856682769728e-06,
      "loss": 0.6039,
      "step": 800
    },
    {
      "epoch": 0.46195652173913043,
      "grad_norm": 15.008636474609375,
      "learning_rate": 9.408212560386473e-06,
      "loss": 0.6314,
      "step": 850
    },
    {
      "epoch": 0.4891304347826087,
      "grad_norm": 6.631000518798828,
      "learning_rate": 9.307568438003222e-06,
      "loss": 0.7266,
      "step": 900
    },
    {
      "epoch": 0.5163043478260869,
      "grad_norm": 9.905613899230957,
      "learning_rate": 9.206924315619969e-06,
      "loss": 0.6624,
      "step": 950
    },
    {
      "epoch": 0.5434782608695652,
      "grad_norm": 13.82048511505127,
      "learning_rate": 9.106280193236716e-06,
      "loss": 0.5971,
      "step": 1000
    },
    {
      "epoch": 0.5706521739130435,
      "grad_norm": 10.97307300567627,
      "learning_rate": 9.005636070853463e-06,
      "loss": 0.6621,
      "step": 1050
    },
    {
      "epoch": 0.5978260869565217,
      "grad_norm": 6.645138263702393,
      "learning_rate": 8.90499194847021e-06,
      "loss": 0.6195,
      "step": 1100
    },
    {
      "epoch": 0.625,
      "grad_norm": 19.04903221130371,
      "learning_rate": 8.804347826086957e-06,
      "loss": 0.5684,
      "step": 1150
    },
    {
      "epoch": 0.6521739130434783,
      "grad_norm": 3.2631289958953857,
      "learning_rate": 8.703703703703705e-06,
      "loss": 0.577,
      "step": 1200
    },
    {
      "epoch": 0.6793478260869565,
      "grad_norm": 5.16421365737915,
      "learning_rate": 8.603059581320452e-06,
      "loss": 0.5752,
      "step": 1250
    },
    {
      "epoch": 0.7065217391304348,
      "grad_norm": 14.548922538757324,
      "learning_rate": 8.502415458937199e-06,
      "loss": 0.5222,
      "step": 1300
    },
    {
      "epoch": 0.7336956521739131,
      "grad_norm": 35.01688003540039,
      "learning_rate": 8.401771336553946e-06,
      "loss": 0.7212,
      "step": 1350
    },
    {
      "epoch": 0.7608695652173914,
      "grad_norm": 12.004179000854492,
      "learning_rate": 8.301127214170693e-06,
      "loss": 0.5907,
      "step": 1400
    },
    {
      "epoch": 0.7880434782608695,
      "grad_norm": 17.919300079345703,
      "learning_rate": 8.20048309178744e-06,
      "loss": 0.5463,
      "step": 1450
    },
    {
      "epoch": 0.8152173913043478,
      "grad_norm": 42.389678955078125,
      "learning_rate": 8.099838969404187e-06,
      "loss": 0.5077,
      "step": 1500
    },
    {
      "epoch": 0.842391304347826,
      "grad_norm": 3.7581164836883545,
      "learning_rate": 7.999194847020934e-06,
      "loss": 0.5113,
      "step": 1550
    },
    {
      "epoch": 0.8695652173913043,
      "grad_norm": 19.534931182861328,
      "learning_rate": 7.898550724637682e-06,
      "loss": 0.552,
      "step": 1600
    },
    {
      "epoch": 0.8967391304347826,
      "grad_norm": 70.55868530273438,
      "learning_rate": 7.797906602254429e-06,
      "loss": 0.5187,
      "step": 1650
    },
    {
      "epoch": 0.9239130434782609,
      "grad_norm": 7.407011032104492,
      "learning_rate": 7.697262479871176e-06,
      "loss": 0.5304,
      "step": 1700
    },
    {
      "epoch": 0.9510869565217391,
      "grad_norm": 17.54728889465332,
      "learning_rate": 7.596618357487924e-06,
      "loss": 0.5928,
      "step": 1750
    },
    {
      "epoch": 0.9782608695652174,
      "grad_norm": 9.951366424560547,
      "learning_rate": 7.49597423510467e-06,
      "loss": 0.509,
      "step": 1800
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.87822777928785,
      "eval_f1_macro": 0.8198928704505658,
      "eval_f1_weighted": 0.8776767663384684,
      "eval_loss": 0.5368145108222961,
      "eval_runtime": 20.0124,
      "eval_samples_per_second": 183.836,
      "eval_steps_per_second": 5.746,
      "step": 1840
    }
  ],
  "logging_steps": 50,
  "max_steps": 5520,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 2.742962313493709e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}