metadata
tags:
- vllm
lm_eval --model vllm --model_args pretrained=/home/mgoin/code/nemotron-3-8b-chat-4k-sft-HF --tasks gsm8k --num_fewshot 5 --batch_size auto
vllm (pretrained=/home/mgoin/code/nemotron-3-8b-chat-4k-sft-HF), gen_kwargs: (None), limit: None, num_fewshot: 5, batch_size: auto
|Tasks|Version| Filter |n-shot| Metric | |Value | |Stderr|
|-----|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|gsm8k| 3|flexible-extract| 5|exact_match|↑ |0.1031|± |0.0084|
| | |strict-match | 5|exact_match|↑ |0.1016|± |0.0083|