Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model:
|
| 3 |
+
- huihui-ai/Qwen3-32B-abliterated
|
| 4 |
+
---
|
| 5 |
+
vllm (pretrained=/root/autodl-tmp/Qwen3-32B-abliterated,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true,tensor_parallel_size=4,gpu_memory_utilization=0.8), 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|↑ |0.900|± |0.0190|
|
| 9 |
+
| | |strict-match | 5|exact_match|↑ |0.896|± |0.0193|
|
| 10 |
+
|
| 11 |
+
vllm (pretrained=/root/autodl-tmp/Qwen3-32B-abliterated,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true,tensor_parallel_size=4,gpu_memory_utilization=0.8), 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|↑ |0.852|± |0.0159|
|
| 15 |
+
| | |strict-match | 5|exact_match|↑ |0.840|± |0.0164|
|
| 16 |
+
|
| 17 |
+
| Groups |Version|Filter|n-shot|Metric| |Value | |Stderr|
|
| 18 |
+
|------------------|------:|------|------|------|---|-----:|---|-----:|
|
| 19 |
+
|mmlu | 2|none | |acc |↑ |0.7988|± |0.0131|
|
| 20 |
+
| - humanities | 2|none | |acc |↑ |0.7897|± |0.0269|
|
| 21 |
+
| - other | 2|none | |acc |↑ |0.7590|± |0.0298|
|
| 22 |
+
| - social sciences| 2|none | |acc |↑ |0.8722|± |0.0252|
|
| 23 |
+
| - stem | 2|none | |acc |↑ |0.7860|± |0.0230|
|
| 24 |
+
|
| 25 |
+
|
| 26 |
+
vllm (pretrained=/root/autodl-tmp/Qwen3-32B-abliterated-awq,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true,tensor_parallel_size=2), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: 1
|
| 27 |
+
|Tasks|Version| Filter |n-shot| Metric | |Value| |Stderr|
|
| 28 |
+
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|
| 29 |
+
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.892|± |0.0197|
|
| 30 |
+
| | |strict-match | 5|exact_match|↑ |0.888|± |0.0200|
|
| 31 |
+
|
| 32 |
+
vllm (pretrained=/root/autodl-tmp/Qwen3-32B-abliterated-awq,add_bos_token=true,max_model_len=3096,dtype=bfloat16,trust_remote_code=true), 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|↑ |0.864|± |0.0153|
|
| 36 |
+
| | |strict-match | 5|exact_match|↑ |0.862|± |0.0154|
|
| 37 |
+
|
| 38 |
+
| Groups |Version|Filter|n-shot|Metric| |Value | |Stderr|
|
| 39 |
+
|------------------|------:|------|------|------|---|-----:|---|-----:|
|
| 40 |
+
|mmlu | 2|none | |acc |↑ |0.7871|± |0.0131|
|
| 41 |
+
| - humanities | 2|none | |acc |↑ |0.8000|± |0.0266|
|
| 42 |
+
| - other | 2|none | |acc |↑ |0.7692|± |0.0280|
|
| 43 |
+
| - social sciences| 2|none | |acc |↑ |0.8611|± |0.0260|
|
| 44 |
+
| - stem | 2|none | |acc |↑ |0.7439|± |0.0240|
|