|
|
--- |
|
|
base_model: |
|
|
- grimjim/Magnolia-v3-Gemma2-8k-9B |
|
|
library_name: transformers |
|
|
tags: |
|
|
- mergekit |
|
|
- merge |
|
|
- roleplay |
|
|
- story-writing |
|
|
- adventure |
|
|
- gemma-2 |
|
|
- rp |
|
|
- nsfw |
|
|
language: |
|
|
- en |
|
|
- zh |
|
|
- ja |
|
|
--- |
|
|
| <img style="float:left;margin-right:0.4em" src="https://qu.ax/gGdYM.webp"> **For RP & story gen,<br/>a nice fine-tuning of Gemma-2-9B could surprise you with some highly creative and authentic expressions way surpassing its size, which Gemma-3-12B even got no match.<br/>Yet the glitches are obvious too, and hard to ignore.<br/>As it's like breaking a perfect sentence with one word so weird<br/>that it may totally come from another language...<br/><br/>Among tons of works trying to stabilize the bitch,<br/>I enjoy [grimjim/Magnolia-v3-Gemma2-8k-9B](https://huggingface.co/grimjim/Magnolia-v3-Gemma2-8k-9B) the most.<br/>So I picked the rich [recoilme/recoilme-gemma-2-9B-v0.2](https://huggingface.co/recoilme/recoilme-gemma-2-9B-v0.2) plus the strong [lemon07r/Gemma-2-Ataraxy-v4d-9B](https://huggingface.co/lemon07r/Gemma-2-Ataraxy-v4d-9B) to tame it with one last merge.<br/>And failed again...<br/><br/>It's just slightly smarter, more sensitive to NSFW directions with a little rebellious tendency.<br/>So keep retrying and editing.<br/>It's 9B, after all.** | |
|
|
|:---:| |
|
|
<small>*"It feels few steps to perfection, 'cause it's google."*</small> |
|
|
```yaml |
|
|
models: |
|
|
- model: grimjim/Magnolia-v3-Gemma2-8k-9B |
|
|
- model: recoilme/recoilme-gemma-2-9B-v0.2 |
|
|
parameters: |
|
|
density: [0.5, 0.7, 0.6, 0.7, 0.5] |
|
|
epsilon: [0.05, 0.07, 0.06, 0.07, 0.05] |
|
|
weight: [-0.01150, 0.01793, -0.01034, 0.01855, -0.01876] |
|
|
- model: lemon07r/Gemma-2-Ataraxy-v4d-9B |
|
|
parameters: |
|
|
density: [0.5, 0.3, 0.4, 0.3, 0.5] |
|
|
epsilon: [0.05, 0.03, 0.04, 0.03, 0.05] |
|
|
weight: [0.01763, -0.01992, 0.01975, -0.01096, 0.01951] |
|
|
merge_method: della |
|
|
base_model: grimjim/Magnolia-v3-Gemma2-8k-9B |
|
|
parameters: |
|
|
normalize: false |
|
|
lambda: 0.66 |
|
|
tokenizer_source: base |
|
|
dtype: bfloat16 |
|
|
``` |
