List your company
ResearchFormerly WizWand

Open-ended Generation on AlpacaEval 1.0

7,904Win RateLLAMA-2-CHAT
-315.121,818.693,952.56,086.31May 23, 2024
Updated 8mo ago

Evaluation results

MethodLinks
LLAMA-2-CHATBackbone=LLAMA-2-7B-BASE2024.057,904
IMBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Less BBH ICL, Number of training examples=13,5332024.054,415
NEFTUNEBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Less BBH ICL, Number of training examples=13,5332024.053,981
ITBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Less BBH ICL, Number of training examples=13,5332024.053,620
ITBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=LIMA, Number of training examples=1,0302024.053,306
IMBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=LIMA, Number of training examples=1,0302024.053,294
NEFTUNEBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Alpagasus Dolly 9k, Number of training examples=9,2292024.053,161
IMBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Alpagasus Dolly 9k, Number of training examples=9,2292024.053,077
NEFTUNEBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=LIMA, Number of training examples=1,0302024.053,051
ITBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Alpagasus Dolly 9k, Number of training examples=9,2292024.052,154
NEFTUNEBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Alpagasus Alpaca 5k, Number of training examples=5,3052024.052,137
IMBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Alpagasus Alpaca 5k, Number of training examples=5,3052024.051,952
ITBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Alpagasus Alpaca 5k, Number of training examples=5,3052024.051,629
IMBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Alpagasus Dolly 3k, Number of training examples=2,9962024.051,511
NEFTUNEBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Alpagasus Dolly 3k, Number of training examples=2,9962024.051,404
ITBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Alpagasus Dolly 3k, Number of training examples=2,9962024.051,342
IMBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Less Tydiqa, Number of training examples=13,5332024.051,010
IMBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Less MMLU Chat, Number of training examples=13,5332024.05978
NEFTUNEBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Less Tydiqa, Number of training examples=13,5332024.05835
NEFTUNEBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Less MMLU Chat, Number of training examples=13,5332024.05622
ITBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Less Tydiqa, Number of training examples=13,5332024.05512
ITBackbone=LLAMA-2-7B-BASE, Instruction Tuning Dataset=Less MMLU Chat, Number of training examples=13,5332024.05442
LLAMA-2-BASEBackbone=LLAMA-2-7B-BASE2024.051