metadata
library_name: peft
license: apache-2.0
base_model: answerdotai/ModernBERT-base
tags:
- generated_from_trainer
model-index:
- name: defiant-crow-853
results: []
defiant-crow-853
This model is a fine-tuned version of answerdotai/ModernBERT-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2274
- Hamming Loss: 0.0824
- Zero One Loss: 0.6800
- Jaccard Score: 0.6658
- Hamming Loss Optimised: 0.0816
- Hamming Loss Threshold: 0.4853
- Zero One Loss Optimised: 0.6275
- Zero One Loss Threshold: 0.2569
- Jaccard Score Optimised: 0.5322
- Jaccard Score Threshold: 0.2068
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: 5.381320896322974e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 2024
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.8546741355228837,0.9482717959895227) and epsilon=1e-07 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | Hamming Loss | Zero One Loss | Jaccard Score | Hamming Loss Optimised | Hamming Loss Threshold | Zero One Loss Optimised | Zero One Loss Threshold | Jaccard Score Optimised | Jaccard Score Threshold |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 100 | 0.3297 | 0.1116 | 0.9938 | 0.9931 | 0.1092 | 0.3475 | 0.7925 | 0.2159 | 0.7520 | 0.1773 |
| No log | 2.0 | 200 | 0.2713 | 0.0934 | 0.8 | 0.7944 | 0.0912 | 0.4406 | 0.7163 | 0.2528 | 0.6303 | 0.1708 |
| No log | 3.0 | 300 | 0.2361 | 0.0846 | 0.7037 | 0.6927 | 0.0839 | 0.4485 | 0.6375 | 0.2889 | 0.5551 | 0.2112 |
| No log | 4.0 | 400 | 0.2274 | 0.0824 | 0.6800 | 0.6658 | 0.0816 | 0.4853 | 0.6275 | 0.2569 | 0.5322 | 0.2068 |
Framework versions
- PEFT 0.13.2
- Transformers 4.48.0.dev0
- Pytorch 2.5.1+cu124
- Datasets 3.1.0
- Tokenizers 0.21.0