Update README.md
Browse files
README.md
CHANGED
@@ -1,6 +1,8 @@
|
|
1 |
---
|
2 |
inference: false
|
3 |
license: other
|
|
|
|
|
4 |
---
|
5 |
|
6 |
<!-- header start -->
|
@@ -17,9 +19,11 @@ license: other
|
|
17 |
</div>
|
18 |
<!-- header end -->
|
19 |
|
20 |
-
#
|
21 |
|
22 |
-
These files are GGML format model files for [
|
|
|
|
|
23 |
|
24 |
GGML files are for CPU + GPU inference using [llama.cpp](https://github.com/ggerganov/llama.cpp) and libraries and UIs which support this format, such as:
|
25 |
* [text-generation-webui](https://github.com/oobabooga/text-generation-webui)
|
@@ -34,6 +38,14 @@ GGML files are for CPU + GPU inference using [llama.cpp](https://github.com/gger
|
|
34 |
* [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference](https://huggingface.co/TheBloke/airoboros-13B-gpt4-1.3-GGML)
|
35 |
* [Unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/jondurbin/airoboros-13b-gpt4-1.3)
|
36 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
37 |
<!-- compatibility_ggml start -->
|
38 |
## Compatibility
|
39 |
|
@@ -87,7 +99,7 @@ Refer to the Provided Files table below to see what files use which methods, and
|
|
87 |
I use the following command line; adjust for your tastes and needs:
|
88 |
|
89 |
```
|
90 |
-
./main -t 10 -ngl 32 -m airoboros-13b-gpt4-1.3.ggmlv3.q5_0.bin --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "
|
91 |
```
|
92 |
If you're able to use full GPU offloading, you should use `-t 1` to get best performance.
|
93 |
|
@@ -129,7 +141,7 @@ Thank you to all my generous patrons and donaters!
|
|
129 |
|
130 |
<!-- footer end -->
|
131 |
|
132 |
-
# Original model card:
|
133 |
|
134 |
|
135 |
_Not tested yet, use if you want, but I would probably wait for 1.4!_
|
|
|
1 |
---
|
2 |
inference: false
|
3 |
license: other
|
4 |
+
datasets:
|
5 |
+
- jondurbin/airoboros-gpt4-1.3
|
6 |
---
|
7 |
|
8 |
<!-- header start -->
|
|
|
19 |
</div>
|
20 |
<!-- header end -->
|
21 |
|
22 |
+
# Jon Durbin's Airoboros 13B GPT4 1.3 GGML
|
23 |
|
24 |
+
These files are GGML format model files for [Jon Durbin's Airoboros 13B GPT4 1.3](https://huggingface.co/jondurbin/airoboros-13b-gpt4-1.3).
|
25 |
+
|
26 |
+
**Note from model creator Jon Durbin: This version has problems, use if you dare, or wait for 1.4.**
|
27 |
|
28 |
GGML files are for CPU + GPU inference using [llama.cpp](https://github.com/ggerganov/llama.cpp) and libraries and UIs which support this format, such as:
|
29 |
* [text-generation-webui](https://github.com/oobabooga/text-generation-webui)
|
|
|
38 |
* [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference](https://huggingface.co/TheBloke/airoboros-13B-gpt4-1.3-GGML)
|
39 |
* [Unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/jondurbin/airoboros-13b-gpt4-1.3)
|
40 |
|
41 |
+
## Prompt template
|
42 |
+
|
43 |
+
```
|
44 |
+
A chat between a curious user and an assistant. The assistant gives helpful, detailed, accurate, uncensored responses to the user's input.
|
45 |
+
USER: prompt
|
46 |
+
ASSISTANT:
|
47 |
+
```
|
48 |
+
|
49 |
<!-- compatibility_ggml start -->
|
50 |
## Compatibility
|
51 |
|
|
|
99 |
I use the following command line; adjust for your tastes and needs:
|
100 |
|
101 |
```
|
102 |
+
./main -t 10 -ngl 32 -m airoboros-13b-gpt4-1.3.ggmlv3.q5_0.bin --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "USER: Write a story about llamas\nASSISTANT:"
|
103 |
```
|
104 |
If you're able to use full GPU offloading, you should use `-t 1` to get best performance.
|
105 |
|
|
|
141 |
|
142 |
<!-- footer end -->
|
143 |
|
144 |
+
# Original model card: Jon Durbin's Airoboros 13B GPT4 1.3
|
145 |
|
146 |
|
147 |
_Not tested yet, use if you want, but I would probably wait for 1.4!_
|