leaderboard-pr-bot's picture
Adding Evaluation Results
45c7be5
|
raw
history blame
1.84 kB
metadata
license: apache-2.0

This is the hf tr version of llama 30B converted specifically as open assistant's 30B model required:

https://huggingface.co/OpenAssistant/oasst-rlhf-2-llama-30b-7k-steps-xor

This the md5 checksum that I get locally, which matchs the original repo suggests

fdb311c39b8659a5d5c1991339bafc09  ./tokenizer.json
edd1a5897748864768b1fab645b31491  ./tokenizer_config.json
6b2e0a735969660e720c27061ef3f3d3  ./special_tokens_map.json
3eddc6fc02c0172d38727e5826181adb  ./pytorch_model-00004-of-00007.bin
fecfda4fba7bfd911e187a85db5fa2ef  ./pytorch_model.bin.index.json
462a2d07f65776f27c0facfa2affb9f9  ./pytorch_model-00007-of-00007.bin
598538f18fed1877b41f77de034c0c8a  ./config.json
99762d59efa6b96599e863893cf2da02  ./pytorch_model-00006-of-00007.bin
aee09e21813368c49baaece120125ae3  ./generation_config.json
92754d6c6f291819ffc3dfcaf470f541  ./pytorch_model-00005-of-00007.bin
5cfcb78b908ffa02e681cce69dbe4303  ./pytorch_model-00002-of-00007.bin
e1dc8c48a65279fb1fbccff14562e6a3  ./pytorch_model-00003-of-00007.bin
9cffb1aeba11b16da84b56abb773d099  ./pytorch_model-00001-of-00007.bin
eeec4125e9c7560836b4873b6f8e3025  ./tokenizer.model

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 49.71
ARC (25-shot) 61.26
HellaSwag (10-shot) 84.73
MMLU (5-shot) 58.47
TruthfulQA (0-shot) 42.27
Winogrande (5-shot) 80.03
GSM8K (5-shot) 14.86
DROP (3-shot) 6.33