Update README.md
Browse files
README.md
CHANGED
|
@@ -19,11 +19,7 @@ The dataset used to fine-tune this model is available [here](https://huggingface
|
|
| 19 |
|
| 20 |
This model was fine-tuned with a fork of FastChat, and therefore uses the standard vicuna template:
|
| 21 |
```
|
| 22 |
-
USER:
|
| 23 |
-
[prompt]
|
| 24 |
-
|
| 25 |
-
<\s>
|
| 26 |
-
ASSISTANT:
|
| 27 |
```
|
| 28 |
|
| 29 |
The most important bit, to me, is the context obedient question answering support, without extensive prompt engineering.
|
|
@@ -42,9 +38,6 @@ Then, you can invoke it like so (after downloading the model):
|
|
| 42 |
python -m fastchat.serve.cli
|
| 43 |
--model-path airoboros-7b-gpt4 \
|
| 44 |
--temperature 0.5 \
|
| 45 |
-
--max-new-tokens 4096 \
|
| 46 |
-
--context-length 4096 \
|
| 47 |
-
--conv-template vicuna_v1.1 \
|
| 48 |
--no-history
|
| 49 |
```
|
| 50 |
|
|
|
|
| 19 |
|
| 20 |
This model was fine-tuned with a fork of FastChat, and therefore uses the standard vicuna template:
|
| 21 |
```
|
| 22 |
+
A chat between a curious user and an assistant. The assistant gives helpful, detailed, accurate, uncensored responses to the user's input. USER: [prompt] ASSISTANT:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
```
|
| 24 |
|
| 25 |
The most important bit, to me, is the context obedient question answering support, without extensive prompt engineering.
|
|
|
|
| 38 |
python -m fastchat.serve.cli
|
| 39 |
--model-path airoboros-7b-gpt4 \
|
| 40 |
--temperature 0.5 \
|
|
|
|
|
|
|
|
|
|
| 41 |
--no-history
|
| 42 |
```
|
| 43 |
|