---
library_name: transformers
license: apache-2.0
base_model: answerdotai/ModernBERT-base
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: modernbert-topics-1m
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# modernbert-topics-1m

This model is a fine-tuned version of [answerdotai/ModernBERT-base](https://huggingface.co/answerdotai/ModernBERT-base) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5923
- Accuracy: 0.7949
- F1: 0.7948
- Worst Group Accuracy: 0.5723

## 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: 5e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 4
- total_train_batch_size: 1024
- total_eval_batch_size: 256
- 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_steps: 5000
- num_epochs: 5

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Accuracy | F1     | Worst Group Accuracy |
|:-------------:|:------:|:----:|:---------------:|:--------:|:------:|:--------------------:|
| 6.7028        | 0.5223 | 500  | 0.8360          | 0.7279   | 0.7269 | 0.4539               |
| 2.9939        | 1.0439 | 1000 | 0.7108          | 0.7635   | 0.7632 | 0.5876               |
| 2.625         | 1.5662 | 1500 | 0.6393          | 0.7785   | 0.7778 | 0.6283               |
| 2.425         | 2.0878 | 2000 | 0.6043          | 0.7886   | 0.7880 | 0.6098               |
| 2.2422        | 2.6101 | 2500 | 0.5870          | 0.7908   | 0.7902 | 0.6272               |
| 2.158         | 3.1316 | 3000 | 0.5723          | 0.7944   | 0.7939 | 0.6427               |
| 1.9898        | 3.6540 | 3500 | 0.5684          | 0.7947   | 0.7945 | 0.6595               |
| 1.8709        | 4.1755 | 4000 | 0.6023          | 0.7941   | 0.7938 | 0.6061               |
| 1.6459        | 4.6978 | 4500 | 0.5923          | 0.7949   | 0.7948 | 0.5723               |


### Framework versions

- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1