Update README.md
Browse files
README.md
CHANGED
@@ -11,6 +11,31 @@ tags: []
|
|
11 |
|
12 |
## Model Details
|
13 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
### Model Description
|
15 |
|
16 |
<!-- Provide a longer summary of what this model is. -->
|
|
|
11 |
|
12 |
## Model Details
|
13 |
|
14 |
+
MODEL_NAME = "Llama-3-8B-Instruct-MoE-4"
|
15 |
+
yaml_config = """
|
16 |
+
base_model: meta-llama/Meta-Llama-3-8B-Instruct
|
17 |
+
gate_mode: random
|
18 |
+
dtype: bfloat16
|
19 |
+
experts_per_token: 2
|
20 |
+
experts:
|
21 |
+
- source_model: meta-llama/Meta-Llama-3-8B-Instruct
|
22 |
+
positive_prompts: []
|
23 |
+
- source_model: meta-llama/Meta-Llama-3-8B-Instruct
|
24 |
+
positive_prompts: []
|
25 |
+
- source_model: meta-llama/Meta-Llama-3-8B-Instruct
|
26 |
+
positive_prompts: []
|
27 |
+
- source_model: meta-llama/Meta-Llama-3-8B-Instruct
|
28 |
+
positive_prompts: []
|
29 |
+
- source_model: meta-llama/Meta-Llama-3-8B-Instruct
|
30 |
+
positive_prompts: []
|
31 |
+
- source_model: meta-llama/Meta-Llama-3-8B-Instruct
|
32 |
+
positive_prompts: []
|
33 |
+
- source_model: meta-llama/Meta-Llama-3-8B-Instruct
|
34 |
+
positive_prompts: []
|
35 |
+
- source_model: meta-llama/Meta-Llama-3-8B-Instruct
|
36 |
+
positive_prompts: []
|
37 |
+
"""
|
38 |
+
|
39 |
### Model Description
|
40 |
|
41 |
<!-- Provide a longer summary of what this model is. -->
|