Update README.md
Browse files
README.md
CHANGED
@@ -39,11 +39,23 @@ Many thanks to William Beauchamp from [Chai](https://chai-research.com/) for pro
|
|
39 |
* [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference](https://huggingface.co/TheBloke/Redmond-Puffin-13B-GGML)
|
40 |
* [Original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/NousResearch/Redmond-Puffin-13B)
|
41 |
|
42 |
-
## Prompt template: Human-
|
43 |
|
44 |
```
|
45 |
-
### human:
|
46 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
47 |
```
|
48 |
|
49 |
## Provided files
|
@@ -131,9 +143,13 @@ model = AutoGPTQForCausalLM.from_quantized(model_name_or_path,
|
|
131 |
"""
|
132 |
|
133 |
prompt = "Tell me about AI"
|
134 |
-
prompt_template=f'''### human:
|
135 |
-
|
136 |
-
|
|
|
|
|
|
|
|
|
137 |
|
138 |
print("\n\n*** Generate:")
|
139 |
|
|
|
39 |
* [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference](https://huggingface.co/TheBloke/Redmond-Puffin-13B-GGML)
|
40 |
* [Original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/NousResearch/Redmond-Puffin-13B)
|
41 |
|
42 |
+
## Prompt template: Human-Response
|
43 |
|
44 |
```
|
45 |
+
### human: {prompt}
|
46 |
+
|
47 |
+
### response:
|
48 |
+
```
|
49 |
+
Optional reccomended pre-prompt / system prompt:
|
50 |
+
|
51 |
+
```
|
52 |
+
### human: Interact in conversation to the best of your ability, please be concise, logical, intelligent and coherent.
|
53 |
+
|
54 |
+
### response: Sure! sounds good.
|
55 |
+
|
56 |
+
### human: {prompt}
|
57 |
+
|
58 |
+
### response:
|
59 |
```
|
60 |
|
61 |
## Provided files
|
|
|
143 |
"""
|
144 |
|
145 |
prompt = "Tell me about AI"
|
146 |
+
prompt_template=f'''### human: Interact in conversation to the best of your ability, please be concise, logical, intelligent and coherent.
|
147 |
+
|
148 |
+
### response: Sure! sounds good.
|
149 |
+
|
150 |
+
### human: {prompt}
|
151 |
+
|
152 |
+
### response:'''
|
153 |
|
154 |
print("\n\n*** Generate:")
|
155 |
|