Commit
•
5973ddf
1
Parent(s):
a28115e
Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model:
|
| 3 |
+
- TouchNight/Ministral-8B-Instruct-2410-HF
|
| 4 |
+
---
|
| 5 |
+
It is worth noting that compared with the prince-canuma version, this version is smaller in size after quantization and its accuracy is also improved by one percentage point.
|
| 6 |
+
|
| 7 |
+
vllm (pretrained=/root/autodl-tmp/Ministral-8B-Instruct-2410-HF,add_bos_token=true,tensor_parallel_size=2,max_model_len=2048,dtype=float16), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto
|
| 8 |
+
|Tasks|Version| Filter |n-shot| Metric | |Value| |Stderr|
|
| 9 |
+
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|
| 10 |
+
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.820|± |0.0243|
|
| 11 |
+
| | |strict-match | 5|exact_match|↑ |0.816|± |0.0246|
|
| 12 |
+
|
| 13 |
+
vllm (pretrained=/root/autodl-tmp/Ministral-8B-Instruct-2410-HF,add_bos_token=true,tensor_parallel_size=2,max_model_len=2048,dtype=bfloat16), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto
|
| 14 |
+
|Tasks|Version| Filter |n-shot| Metric | |Value| |Stderr|
|
| 15 |
+
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|
| 16 |
+
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.804|± |0.0252|
|
| 17 |
+
| | |strict-match | 5|exact_match|↑ |0.804|± |0.0252|
|
| 18 |
+
|
| 19 |
+
vllm (pretrained=/root/autodl-tmp/Ministral-8B-Instruct-2410-HF,add_bos_token=true,tensor_parallel_size=2,max_model_len=2048,dtype=float32), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto
|
| 20 |
+
|Tasks|Version| Filter |n-shot| Metric | |Value| |Stderr|
|
| 21 |
+
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|
| 22 |
+
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.820|± |0.0243|
|
| 23 |
+
| | |strict-match | 5|exact_match|↑ |0.816|± |0.0246|
|
| 24 |
+
|
| 25 |
+
vllm (pretrained=/root/autodl-tmp/output,add_bos_token=true,tensor_parallel_size=2,max_model_len=2048,dtype=float16), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto
|
| 26 |
+
|Tasks|Version| Filter |n-shot| Metric | |Value| |Stderr|
|
| 27 |
+
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|
| 28 |
+
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.816|± |0.0246|
|
| 29 |
+
| | |strict-match | 5|exact_match|↑ |0.812|± |0.0248|
|
| 30 |
+
|
| 31 |
+
vllm (pretrained=/root/autodl-tmp/output,add_bos_token=true,tensor_parallel_size=2,max_model_len=2048,dtype=bfloat16), gen_kwargs: (None), limit: 250.0, num_fewshot: 5, batch_size: auto
|
| 32 |
+
|Tasks|Version| Filter |n-shot| Metric | |Value| |Stderr|
|
| 33 |
+
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|
| 34 |
+
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.796|± |0.0255|
|
| 35 |
+
| | |strict-match | 5|exact_match|↑ |0.792|± |0.0257|
|