File size: 1,576 Bytes
2830b44 ea3b28f 2830b44 0b2503b a3edbde 38ae4df 4faccd0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 |
---
license: apache-2.0
tags:
- merge
---
This model is a [Slerp Merge](https://github.com/cg123/mergekit/blob/main/mergekit/merge_methods/slerp.py) of [cookinai/CatMacaroni-Slerp](https://huggingface.co/cookinai/CatMacaroni-Slerp) and [mncai/mistral-7b-dpo-v5](https://huggingface.co/mncai/mistral-7b-dpo-v5).
# Evaluation Results
### HuggingFace Leaderboard
| Average | ARC | HellaSwag | MMLU | TruthfulQA | Winogrande | GSM8K |
| --- | --- | --- | --- | --- | --- | --- |
| 73.1 | 69.62 | 87.09 | 64.81 | 62.82 | 81.45 | 72.78 |
The model did achieve an improvement in TruthfulQA over `cookinai/CatMacaroni-Slerp` and GSM8K over `mncai/mistral-7b-dpo-v5`
which was the goal of the merge leading to an average score that was a better than both. It is unclear why the TruthfulQA metric
is still meaningfully lower than the base `mncai/mistral-7b-dpo-v5`.
# Training Details
.yaml file for mergekit
```yaml
slices:
- sources:
- model: cookinai/CatMacaroni-Slerp
layer_range: [0, 32]
- model: mncai/mistral-7b-dpo-v5
layer_range: [0, 32]
merge_method: slerp
base_model: mncai/mistral-7b-dpo-v5
parameters:
t:
- filter: self_attn
value: [0, 0.5, 0.3, 0.7, 1]
- filter: mlp
value: [1, 0.5, 0.7, 0.3, 0]
- value: 0.5 # fallback for rest of tensors
dtype: float16
```
# Bias, Risks, and Limitations
The model has not been evaluated for safety and is only intended for research and experiments. |