Update README.md
Browse files
README.md
CHANGED
|
@@ -5,16 +5,18 @@ library_name: transformers
|
|
| 5 |
tags:
|
| 6 |
- mergekit
|
| 7 |
- merge
|
| 8 |
-
|
| 9 |
---
|
| 10 |
-
#
|
| 11 |
|
| 12 |
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
| 13 |
|
| 14 |
## Merge Details
|
| 15 |
### Merge Method
|
| 16 |
|
| 17 |
-
This model was merged using the passthrough merge method
|
|
|
|
|
|
|
| 18 |
|
| 19 |
### Models Merged
|
| 20 |
|
|
@@ -55,4 +57,4 @@ slices:
|
|
| 55 |
- sources:
|
| 56 |
- layer_range: [24, 32]
|
| 57 |
model: meta-llama/Meta-Llama-3-8B-Instruct
|
| 58 |
-
```
|
|
|
|
| 5 |
tags:
|
| 6 |
- mergekit
|
| 7 |
- merge
|
| 8 |
+
license: llama3
|
| 9 |
---
|
| 10 |
+
# Llama-3-15B-Instruct-zeroed
|
| 11 |
|
| 12 |
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
| 13 |
|
| 14 |
## Merge Details
|
| 15 |
### Merge Method
|
| 16 |
|
| 17 |
+
This model was merged using the passthrough merge method while zeroing `o_proj` and `down_proj` which led to an decrease in perplexity (good)
|
| 18 |
+
compared to similar 15B merges. This was a recommendation from [Charles Goddard](https://huggingface.co/chargoddard) - thank you for sharing the method of merging as well as Toasty
|
| 19 |
+
Pigeon for bringing it to my attention!
|
| 20 |
|
| 21 |
### Models Merged
|
| 22 |
|
|
|
|
| 57 |
- sources:
|
| 58 |
- layer_range: [24, 32]
|
| 59 |
model: meta-llama/Meta-Llama-3-8B-Instruct
|
| 60 |
+
```
|