Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,66 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
base_model:
|
3 |
+
- huihui-ai/QwQ-32B-abliterated
|
4 |
+
---
|
5 |
+
vllm (pretrained=/root/autodl-tmp/QwQ-32B,add_bos_token=true,max_model_len=4096,dtype=bfloat16), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto
|
6 |
+
|Tasks|Version| Filter |n-shot| Metric | |Value| |Stderr|
|
7 |
+
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|
8 |
+
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.432|± |0.0314|
|
9 |
+
| | |strict-match | 5|exact_match|↑ |0.744|± |0.0277|
|
10 |
+
|
11 |
+
vllm (pretrained=/root/autodl-tmp/QwQ-32B,add_bos_token=true,max_model_len=4096,dtype=bfloat16), gen_kwargs: (None), limit: 500.0, num_fewshot: 5, batch_size: auto
|
12 |
+
|Tasks|Version| Filter |n-shot| Metric | |Value| |Stderr|
|
13 |
+
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|
14 |
+
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.444|± |0.0222|
|
15 |
+
| | |strict-match | 5|exact_match|↑ |0.716|± |0.0202|
|
16 |
+
|
17 |
+
| Groups |Version|Filter|n-shot|Metric| |Value | |Stderr|
|
18 |
+
|------------------|------:|------|------|------|---|-----:|---|-----:|
|
19 |
+
|mmlu | 2|none | |acc |↑ |0.8140|± |0.0125|
|
20 |
+
| - humanities | 2|none | |acc |↑ |0.8359|± |0.0251|
|
21 |
+
| - other | 2|none | |acc |↑ |0.8103|± |0.0269|
|
22 |
+
| - social sciences| 2|none | |acc |↑ |0.8889|± |0.0222|
|
23 |
+
| - stem | 2|none | |acc |↑ |0.7544|± |0.0238|
|
24 |
+
|
25 |
+
|
26 |
+
vllm (pretrained=/root/autodl-tmp/QwQ-32B-abliterated,add_bos_token=true,max_model_len=4096,dtype=bfloat16), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto
|
27 |
+
|Tasks|Version| Filter |n-shot| Metric | |Value| |Stderr|
|
28 |
+
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|
29 |
+
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.528|± |0.0316|
|
30 |
+
| | |strict-match | 5|exact_match|↑ |0.740|± |0.0278|
|
31 |
+
|
32 |
+
vllm (pretrained=/root/autodl-tmp/QwQ-32B-abliterated,add_bos_token=true,max_model_len=4096,dtype=bfloat16), gen_kwargs: (None), limit: 500.0, num_fewshot: 5, batch_size: auto
|
33 |
+
|Tasks|Version| Filter |n-shot| Metric | |Value| |Stderr|
|
34 |
+
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|
35 |
+
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.492|± |0.0224|
|
36 |
+
| | |strict-match | 5|exact_match|↑ |0.742|± |0.0196|
|
37 |
+
|
38 |
+
vllm (pretrained=/root/autodl-tmp/QwQ-32B-abliterated,add_bos_token=true,max_model_len=4096,dtype=bfloat16,max_num_seqs=3), gen_kwargs: (None), limit: 15.0, num_fewshot: None, batch_size: 1
|
39 |
+
| Groups |Version|Filter|n-shot|Metric| |Value | |Stderr|
|
40 |
+
|------------------|------:|------|------|------|---|-----:|---|-----:|
|
41 |
+
|mmlu | 2|none | |acc |↑ |0.8152|± |0.0126|
|
42 |
+
| - humanities | 2|none | |acc |↑ |0.8359|± |0.0253|
|
43 |
+
| - other | 2|none | |acc |↑ |0.8000|± |0.0276|
|
44 |
+
| - social sciences| 2|none | |acc |↑ |0.8722|± |0.0240|
|
45 |
+
| - stem | 2|none | |acc |↑ |0.7754|± |0.0232|
|
46 |
+
|
47 |
+
|
48 |
+
vllm (pretrained=/root/autodl-tmp/QwQ-32B-abliterated-awq,add_bos_token=true,max_model_len=4096,dtype=bfloat16), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto
|
49 |
+
|Tasks|Version| Filter |n-shot| Metric | |Value| |Stderr|
|
50 |
+
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|
51 |
+
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.476|± |0.0316|
|
52 |
+
| | |strict-match | 5|exact_match|↑ |0.752|± |0.0274|
|
53 |
+
|
54 |
+
vllm (pretrained=/root/autodl-tmp/QwQ-32B-abliterated-awq,add_bos_token=true,max_model_len=4096,dtype=bfloat16), gen_kwargs: (None), limit: 500.0, num_fewshot: 5, batch_size: auto
|
55 |
+
|Tasks|Version| Filter |n-shot| Metric | |Value| |Stderr|
|
56 |
+
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|
57 |
+
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.524|± |0.0224|
|
58 |
+
| | |strict-match | 5|exact_match|↑ |0.716|± |0.0202|
|
59 |
+
|
60 |
+
| Groups |Version|Filter|n-shot|Metric| |Value | |Stderr|
|
61 |
+
|------------------|------:|------|------|------|---|-----:|---|-----:|
|
62 |
+
|mmlu | 2|none | |acc |↑ |0.8023|± |0.0130|
|
63 |
+
| - humanities | 2|none | |acc |↑ |0.8000|± |0.0266|
|
64 |
+
| - other | 2|none | |acc |↑ |0.7949|± |0.0284|
|
65 |
+
| - social sciences| 2|none | |acc |↑ |0.8500|± |0.0258|
|
66 |
+
| - stem | 2|none | |acc |↑ |0.7789|± |0.0235|
|