Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,66 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
---
|
| 3 |
+
|
| 4 |
+
library_name: transformers
|
| 5 |
+
tags:
|
| 6 |
+
- mergekit
|
| 7 |
+
- merge
|
| 8 |
+
base_model:
|
| 9 |
+
- bluuwhale/L3-SAO-MIX-8B-V1
|
| 10 |
+
- Sao10K/L3-8B-Niitama-v1
|
| 11 |
+
- Sao10K/L3-8B-Lunaris-v1
|
| 12 |
+
- Sao10K/L3-8B-Tamamo-v1
|
| 13 |
+
- Sao10K/L3-8B-Stheno-v3.2
|
| 14 |
+
|
| 15 |
+
---
|
| 16 |
+
|
| 17 |
+
[](https://hf.co/QuantFactory)
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
# QuantFactory/L3-bluuwhale-SAO-MIX-8B-V1_fp32-merge-calc-GGUF
|
| 21 |
+
This is quantized version of [Casual-Autopsy/L3-bluuwhale-SAO-MIX-8B-V1_fp32-merge-calc](https://huggingface.co/Casual-Autopsy/L3-bluuwhale-SAO-MIX-8B-V1_fp32-merge-calc) created using llama.cpp
|
| 22 |
+
|
| 23 |
+
# Original Model Card
|
| 24 |
+
|
| 25 |
+
# L3-bluuwhale-SAO-MIX-8B-V1_fp32-merge-calc
|
| 26 |
+
|
| 27 |
+
This is a remerge of [bluuwhale's merger](https://huggingface.co/bluuwhale/L3-SAO-MIX-8B-V1) using the exact yaml config with the only difference being that merge calculations are done in fp32 instead of bfp16
|
| 28 |
+
|
| 29 |
+
I've done this since I'm planning to use this for another merger, but you can use as is if you wish.
|
| 30 |
+
|
| 31 |
+
## Merge Details
|
| 32 |
+
### Merge Method
|
| 33 |
+
|
| 34 |
+
This model was merged using the della merge method using [Sao10K/L3-8B-Niitama-v1](https://huggingface.co/Sao10K/L3-8B-Niitama-v1) as a base.
|
| 35 |
+
|
| 36 |
+
### Models Merged
|
| 37 |
+
|
| 38 |
+
The following models were included in the merge:
|
| 39 |
+
* [Sao10K/L3-8B-Lunaris-v1](https://huggingface.co/Sao10K/L3-8B-Lunaris-v1)
|
| 40 |
+
* [Sao10K/L3-8B-Stheno-v3.2](https://huggingface.co/Sao10K/L3-8B-Stheno-v3.2)
|
| 41 |
+
* [Sao10K/L3-8B-Tamamo-v1](https://huggingface.co/Sao10K/L3-8B-Tamamo-v1)
|
| 42 |
+
|
| 43 |
+
### Configuration
|
| 44 |
+
|
| 45 |
+
The following YAML configuration was used to produce this model:
|
| 46 |
+
|
| 47 |
+
```yaml
|
| 48 |
+
models:
|
| 49 |
+
- model: Sao10K/L3-8B-Lunaris-v1
|
| 50 |
+
parameters:
|
| 51 |
+
weight: 1.0
|
| 52 |
+
- model: Sao10K/L3-8B-Stheno-v3.2
|
| 53 |
+
parameters:
|
| 54 |
+
weight: 1.0
|
| 55 |
+
- model: Sao10K/L3-8B-Niitama-v1
|
| 56 |
+
parameters:
|
| 57 |
+
weight: 1.0
|
| 58 |
+
- model: Sao10K/L3-8B-Tamamo-v1
|
| 59 |
+
parameters:
|
| 60 |
+
weight: 1.0
|
| 61 |
+
base_model: Sao10K/L3-8B-Niitama-v1
|
| 62 |
+
merge_method: della
|
| 63 |
+
dtype: float32
|
| 64 |
+
out_dtype: bfloat16
|
| 65 |
+
```
|
| 66 |
+
|