77 lines
1.8 KiB
Markdown
77 lines
1.8 KiB
Markdown
|
---
|
||
|
|
base_model:
|
||
|
|
- mistralai/Mistral-Nemo-Instruct-2407
|
||
|
|
- Sorihon/Geodesic-Phantom-12B-Heretic
|
||
|
|
- Sorihon/Nether-Moon-12B-Heretic
|
||
|
|
- IIEleven11/Kalypso
|
||
|
|
- EldritchLabs/Human-Like-Mistral-Nemo-Instruct-2407-MPOA
|
||
|
|
- EldritchLabs/MN-12B-RP-Ink-Longform-MPOA
|
||
|
|
- MuXodious/Rocinante-X-12B-v1-absolute-heresy
|
||
|
|
- PocketDoc/Dans-DangerousWinds-V1.1.0-12b
|
||
|
|
datasets:
|
||
|
|
- OccultAI/illuminati_imatrix_v1
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
library_name: transformers
|
||
|
|
license: apache-2.0
|
||
|
|
tags:
|
||
|
|
- merge
|
||
|
|
- mergekit
|
||
|
|
---
|
||
|
|
|
||
|
|
# 🚣 Styx 12B
|
||
|
|
|
||
|
|
A saved prototype of an earlier merge test, made before Savage Sands.
|
||
|
|
|
||
|
|
This model uses pre-ablated components and may have some refusals.
|
||
|
|
|
||
|
|
```yaml
|
||
|
|
architecture: MistralForCausalLM
|
||
|
|
base_model: B:/12B/mistralai--Mistral-Nemo-Instruct-2407
|
||
|
|
models:
|
||
|
|
- model: B:/12B/Sorihon--Geodesic-Phantom-12B-Heretic
|
||
|
|
parameters:
|
||
|
|
weight: 0.3
|
||
|
|
density: 0.9
|
||
|
|
epsilon: 0.09
|
||
|
|
- model: B:/12B/Sorihon--Nether-Moon-12B-Heretic
|
||
|
|
parameters:
|
||
|
|
weight: 0.3
|
||
|
|
density: 0.9
|
||
|
|
epsilon: 0.09
|
||
|
|
- model: B:/12B/IIEleven11--Kalypso
|
||
|
|
parameters:
|
||
|
|
weight: 0.3
|
||
|
|
density: 0.9
|
||
|
|
epsilon: 0.09
|
||
|
|
- model: B:/12B/EldritchLabs--Human-Like-Mistral-Nemo-Instruct-2407-MPOA
|
||
|
|
parameters:
|
||
|
|
weight: 0.3
|
||
|
|
density: 0.9
|
||
|
|
epsilon: 0.09
|
||
|
|
- model: B:/12B/EldritchLabs--MN-12B-RP-Ink-Longform-MPOA
|
||
|
|
parameters:
|
||
|
|
weight: 0.3
|
||
|
|
density: 0.9
|
||
|
|
epsilon: 0.09
|
||
|
|
- model: B:/12B/MuXodious-Rocinante-X-12B-v1-absolute-heresy
|
||
|
|
parameters:
|
||
|
|
weight: 0.3
|
||
|
|
density: 0.9
|
||
|
|
epsilon: 0.09
|
||
|
|
- model: B:/12B/PocketDoc--Dans-DangerousWinds-V1.1.0-12b
|
||
|
|
parameters:
|
||
|
|
weight: 0.3
|
||
|
|
density: 0.9
|
||
|
|
epsilon: 0.09
|
||
|
|
merge_method: della
|
||
|
|
parameters:
|
||
|
|
lambda: 1.0
|
||
|
|
normalize: false
|
||
|
|
int8_mask: false
|
||
|
|
rescale: true
|
||
|
|
dtype: float32
|
||
|
|
out_dtype: bfloat16
|
||
|
|
tokenizer:
|
||
|
|
source: B:/12B/Sorihon--Geodesic-Phantom-12B-Heretic
|
||
|
|
```
|