Update README.md
Browse files
README.md
CHANGED
@@ -5,4 +5,29 @@ license_link: https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct/raw/mai
|
|
5 |
---
|
6 |
Merged [cognitivecomputations/dolphin-2.9-llama3-8b](https://huggingface.co/cognitivecomputations/dolphin-2.9-llama3-8b) and [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) because I thought the Dolphin finetune was a bit too 'robot-y' in the answers.
|
7 |
|
8 |
-
GGUF files can be found here: [RDson/Dolphin-less-Llama-3-Instruct-8B-GGUF](https://huggingface.co/RDson/Dolphin-less-Llama-3-Instruct-8B-GGUF).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
---
|
6 |
Merged [cognitivecomputations/dolphin-2.9-llama3-8b](https://huggingface.co/cognitivecomputations/dolphin-2.9-llama3-8b) and [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) because I thought the Dolphin finetune was a bit too 'robot-y' in the answers.
|
7 |
|
8 |
+
GGUF files can be found here: [RDson/Dolphin-less-Llama-3-Instruct-8B-GGUF](https://huggingface.co/RDson/Dolphin-less-Llama-3-Instruct-8B-GGUF).
|
9 |
+
|
10 |
+
Mergekit yaml:
|
11 |
+
```
|
12 |
+
tokenizer_source: union
|
13 |
+
slices:
|
14 |
+
- sources:
|
15 |
+
- model: ollama/llama3/Meta-Llama-3-8B-Instruct
|
16 |
+
layer_range: [0, 32]
|
17 |
+
- model: dolphin-2.9-llama3-8b
|
18 |
+
layer_range: [0, 32]
|
19 |
+
parameters:
|
20 |
+
weight: 0.75
|
21 |
+
merge_method: slerp
|
22 |
+
base_model: ollama/llama3/Meta-Llama-3-8B-Instruct
|
23 |
+
parameters:
|
24 |
+
normalize: true
|
25 |
+
embed_slerp: true
|
26 |
+
t:
|
27 |
+
- filter: self_attn
|
28 |
+
value: [0, 0.5, 0.3, 0.7, 1]
|
29 |
+
- filter: mlp
|
30 |
+
value: [1, 0.5, 0.7, 0.3, 0]
|
31 |
+
- value: 0.5
|
32 |
+
dtype: bfloat16
|
33 |
+
```
|