README.md

metadata

library_name: transformers
license: apache-2.0
base_model: answerdotai/ModernBERT-base
tags:
  - generated_from_trainer
model-index:
  - name: nci-technique-classifier-v5.2
    results: []

nci-technique-classifier-v5.2

This model is a fine-tuned version of answerdotai/ModernBERT-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0173
• Micro F1: 0.7718
• Macro F1: 0.5789

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Micro F1	Macro F1
0.0275	0.1570	200	0.0272	0.6634	0.2831
0.0256	0.3140	400	0.0238	0.6844	0.3147
0.0211	0.4710	600	0.0226	0.7276	0.2792
0.0224	0.6279	800	0.0206	0.7140	0.4159
0.0198	0.7849	1000	0.0203	0.7180	0.4403
0.0175	0.9419	1200	0.0192	0.7481	0.4333
0.018	1.0989	1400	0.0190	0.7320	0.4845
0.017	1.2559	1600	0.0191	0.7199	0.4723
0.0165	1.4129	1800	0.0188	0.7597	0.4633
0.0165	1.5699	2000	0.0182	0.7434	0.5247
0.0167	1.7268	2200	0.0183	0.7345	0.5005
0.0167	1.8838	2400	0.0182	0.7629	0.5162
0.0143	2.0408	2600	0.0180	0.7493	0.5557
0.016	2.1978	2800	0.0183	0.7588	0.5513
0.0157	2.3548	3000	0.0185	0.7663	0.5457
0.0157	2.5118	3200	0.0183	0.7665	0.5756
0.0146	2.6688	3400	0.0179	0.7641	0.5885
0.0123	2.8257	3600	0.0182	0.7719	0.5734
0.0136	2.9827	3800	0.0179	0.7682	0.5952

Framework versions

• Transformers 4.57.3
• Pytorch 2.9.1+cu128
• Datasets 4.4.1
• Tokenizers 0.22.1