Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Instruction Following on Dolly
Loading...
71.3
Score
Ours 12
20.756
33.878
47
60.122
Oct 5, 2023
Jan 1, 2024
Mar 30, 2024
Jun 27, 2024
Sep 24, 2024
Dec 22, 2024
Mar 21, 2025
Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Score
Ours 12
Distillation Method=Ra...
2025.03
71.3
FullKD
Distillation Method=Fu...
2025.03
66.1
Top-K 50
Distillation Method=To...
2025.03
65.4
CE
Distillation Method=No...
2025.03
64.2
Top-K 12
Distillation Method=To...
2025.03
59
LumiNet
Model Architecture=GPT...
2023.10
28.6
LumiNet
Model Architecture=GPT...
2023.10
27.8
Teacher
Model Architecture=Tea...
2023.10
27.6
KD
Model Architecture=GPT...
2023.10
25.9
SeqKD
Model Architecture=GPT...
2023.10
25.6
SFT w/o KD
Model Architecture=GPT...
2023.10
25.5
SFT w/o KD
Model Architecture=GPT...
2023.10
25.4
SeqKD
Model Architecture=GPT...
2023.10
25.3
KD
Model Architecture=GPT...
2023.10
25
LumiNet
Model Architecture=GPT...
2023.10
23.8
SFT w/o KD
Model Architecture=GPT...
2023.10
23.3
KD
Model Architecture=GPT...
2023.10
22.8
SeqKD
Model Architecture=GPT...
2023.10
22.7
Feedback
Search any
task
Search any
task