Update README.md
Browse files
README.md
CHANGED
@@ -43,6 +43,13 @@ The primary use of Dromedary is research on the alignment of large language mode
|
|
43 |
**Primary intended users:**
|
44 |
The primary intended users of the model are researchers in artificial intelligence.
|
45 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
46 |
## Training dataset
|
47 |
Fewer than 300 lines of human annotations (including < 200 seed prompts, 16 generic principles, and 5 exemplars for in-context learning),
|
48 |
|
|
|
43 |
**Primary intended users:**
|
44 |
The primary intended users of the model are researchers in artificial intelligence.
|
45 |
|
46 |
+
## Delta weights
|
47 |
+
We use the following configuration for the LoRA weights:
|
48 |
+
```
|
49 |
+
--lora_target_modules='[q_proj,k_proj,v_proj,o_proj]' \
|
50 |
+
--lora_r=16 \
|
51 |
+
```
|
52 |
+
|
53 |
## Training dataset
|
54 |
Fewer than 300 lines of human annotations (including < 200 seed prompts, 16 generic principles, and 5 exemplars for in-context learning),
|
55 |
|