---
base_model:
- rirv938/gpt2_sequence_classification_base
- rirv938/reward_gpt2_preference_24m_e2
- ChaiML/reward_models_100_170000000_cp_332032
library_name: transformers
tags:
- mergekit
- merge

---
# gpt2_ties_merge_ab_with_classic_e2_d99

This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).

## Merge Details
### Merge Method

This model was merged using the [TIES](https://arxiv.org/abs/2306.01708) merge method using [rirv938/gpt2_sequence_classification_base](https://huggingface.co/rirv938/gpt2_sequence_classification_base) as a base.

### Models Merged

The following models were included in the merge:
* [rirv938/reward_gpt2_preference_24m_e2](https://huggingface.co/rirv938/reward_gpt2_preference_24m_e2)
* [ChaiML/reward_models_100_170000000_cp_332032](https://huggingface.co/ChaiML/reward_models_100_170000000_cp_332032)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
models:
  - model: rirv938/gpt2_sequence_classification_base
    # no parameters necessary for base model
  - model: ChaiML/reward_models_100_170000000_cp_332032
    parameters:
      density: 0.9
      weight: 0.5
  - model: rirv938/reward_gpt2_preference_24m_e2
    parameters:
      density: 0.9
      weight: 0.5
merge_method: ties
base_model: rirv938/gpt2_sequence_classification_base
parameters:
  normalize: true
dtype: float16

```