Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model:
|
| 3 |
+
- huihui-ai/Qwen2.5-14B-Instruct-1M-abliterated
|
| 4 |
+
---
|
| 5 |
+
vllm (pretrained=/root/autodl-tmp/Qwen2.5-14B-Instruct-1M-abliterated,add_bos_token=true,max_model_len=2048,tensor_parallel_size=2,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|↑ |0.868|± |0.0215|
|
| 9 |
+
| | |strict-match | 5|exact_match|↑ |0.872|± |0.0212|
|
| 10 |
+
|
| 11 |
+
vllm (pretrained=/root/autodl-tmp/Qwen2.5-14B-Instruct-1M-abliterated,add_bos_token=true,max_model_len=2048,tensor_parallel_size=2,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|↑ |0.872|± |0.0150|
|
| 15 |
+
| | |strict-match | 5|exact_match|↑ |0.870|± |0.0151|
|
| 16 |
+
|
| 17 |
+
vllm (pretrained=/root/autodl-tmp/Qwen2.5-14B-Instruct-1M-abliterated,add_bos_token=true,max_model_len=700,tensor_parallel_size=2,dtype=bfloat16,enforce_eager=True), gen_kwargs: (None), limit: 7.0, num_fewshot: None, batch_size: 1
|
| 18 |
+
| Groups |Version|Filter|n-shot|Metric| |Value | |Stderr|
|
| 19 |
+
|------------------|------:|------|------|------|---|-----:|---|-----:|
|
| 20 |
+
|mmlu | 2|none | |acc |↑ |0.7769|± |0.0202|
|
| 21 |
+
| - humanities | 2|none | |acc |↑ |0.7692|± |0.0440|
|
| 22 |
+
| - other | 2|none | |acc |↑ |0.7582|± |0.0406|
|
| 23 |
+
| - social sciences| 2|none | |acc |↑ |0.8452|± |0.0376|
|
| 24 |
+
| - stem | 2|none | |acc |↑ |0.7519|± |0.0376|
|
| 25 |
+
|
| 26 |
+
|
| 27 |
+
vllm (pretrained=/root/autodl-tmp/Qwen2.5-14B-Instruct-1M-abliterated-87,add_bos_token=true,max_model_len=2048,tensor_parallel_size=2,dtype=bfloat16), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto
|
| 28 |
+
|Tasks|Version| Filter |n-shot| Metric | |Value| |Stderr|
|
| 29 |
+
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|
| 30 |
+
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.864|± |0.0217|
|
| 31 |
+
| | |strict-match | 5|exact_match|↑ |0.864|± |0.0217|
|
| 32 |
+
|
| 33 |
+
vllm (pretrained=/root/autodl-tmp/Qwen2.5-14B-Instruct-1M-abliterated-87,add_bos_token=true,max_model_len=2048,tensor_parallel_size=2,dtype=bfloat16), gen_kwargs: (None), limit: 500.0, num_fewshot: 5, batch_size: auto
|
| 34 |
+
|Tasks|Version| Filter |n-shot| Metric | |Value| |Stderr|
|
| 35 |
+
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|
| 36 |
+
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.882|± |0.0144|
|
| 37 |
+
| | |strict-match | 5|exact_match|↑ |0.874|± |0.0149|
|
| 38 |
+
|
| 39 |
+
vllm (pretrained=/root/autodl-tmp/Qwen2.5-14B-Instruct-1M-abliterated-87,add_bos_token=true,max_model_len=700,tensor_parallel_size=2,dtype=bfloat16,enforce_eager=True), gen_kwargs: (None), limit: 7.0, num_fewshot: None, batch_size: 1
|
| 40 |
+
| Groups |Version|Filter|n-shot|Metric| |Value | |Stderr|
|
| 41 |
+
|------------------|------:|------|------|------|---|-----:|---|-----:|
|
| 42 |
+
|mmlu | 2|none | |acc |↑ |0.7769|± |0.0201|
|
| 43 |
+
| - humanities | 2|none | |acc |↑ |0.7692|± |0.0440|
|
| 44 |
+
| - other | 2|none | |acc |↑ |0.7692|± |0.0391|
|
| 45 |
+
| - social sciences| 2|none | |acc |↑ |0.8333|± |0.0370|
|
| 46 |
+
| - stem | 2|none | |acc |↑ |0.7519|± |0.0381|
|