eryk-mazus
commited on
Commit
•
edee2d8
1
Parent(s):
75af1c8
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
base_model: eryk-mazus/polka-1.1b-chat
|
3 |
+
inference: false
|
4 |
+
language:
|
5 |
+
- pl
|
6 |
+
license: apache-2.0
|
7 |
+
model_name: Polka-1.1B-Chat
|
8 |
+
model_type: tinyllama
|
9 |
+
model_creator: Eryk Mazuś
|
10 |
+
prompt_template: '<|im_start|>system
|
11 |
+
|
12 |
+
{system_message}<|im_end|>
|
13 |
+
|
14 |
+
<|im_start|>user
|
15 |
+
|
16 |
+
{prompt}<|im_end|>
|
17 |
+
|
18 |
+
<|im_start|>assistant
|
19 |
+
|
20 |
+
'
|
21 |
+
---
|
22 |
+
|
23 |
+
## Prompt template: ChatML
|
24 |
+
|
25 |
+
```
|
26 |
+
<|im_start|>system
|
27 |
+
Jesteś pomocnym asystentem.<|im_end|>
|
28 |
+
<|im_start|>user
|
29 |
+
{prompt}<|im_end|>
|
30 |
+
<|im_start|>assistant
|
31 |
+
|
32 |
+
```
|
33 |
+
|
34 |
+
## Example `llama.cpp` command
|
35 |
+
|
36 |
+
```shell
|
37 |
+
./main -m ./polka-1.1b-chat-gguf/polka-1.1b-chat-Q8_0.gguf --color -c 2048 --temp 0.2 --repeat_penalty 1.1 -n -1 -p "<|im_start|>system\nJesteś pomocnym asystentem.<|im_end|>\n<|im_start|>user\n{prompt}<|im_end|>\n<|im_start|>assistant"
|
38 |
+
```
|
39 |
+
|
40 |
+
Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
|
41 |
+
|
42 |
+
Change `-c 2048` to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically.
|
43 |
+
|
44 |
+
If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
|