Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Commonsense Reasoning on ARC Challenge (test)
Loading...
57.8
Accuracy
LoRA
40.5672
45.0411
49.515
53.9889
May 28, 2025
May 30, 2025
Jun 1, 2025
Jun 3, 2025
Jun 5, 2025
Jun 7, 2025
Jun 10, 2025
Accuracy
Updated 23d ago
Evaluation Results
Method
Method
Links
Accuracy
LoRA
Backbone=Llama-2-13b,...
2025.06
57.8
LoRA
Backbone=Llama-2-13b,...
2025.06
57.6
LoTR
Backbone=Llama-2-13b,...
2025.06
57
MetaTT-4D
Backbone=Llama-2-13b,...
2025.06
56.9
LoTR
Backbone=Llama-2-13b,...
2025.06
55.5
MetaTT-4D
Backbone=Llama-2-13b,...
2025.06
55.2
LoRA
Backbone=Llama-2-7b, R...
2025.06
53.3
VeRA
Backbone=Llama-2-13b,...
2025.06
53.2
LoRA
Backbone=Llama-2-7b, R...
2025.06
52.6
LoTR
Backbone=Llama-2-7b, R...
2025.06
52.1
MetaTT-4D
Backbone=Llama-2-7b, R...
2025.06
51.8
LoTR
Backbone=Llama-2-7b, R...
2025.06
51.2
MetaTT-4D
Backbone=Llama-2-7b, R...
2025.06
50.2
Zero Shot
Backbone=Llama-2-13b,...
2025.06
48.9
VeRA
Backbone=Llama-2-7b, R...
2025.06
47.8
Zero Shot
Backbone=Llama-2-7b, P...
2025.06
46.5
BAM SSMax
Model Size=1B, Activat...
2025.05
41.32
RoPE SSMax
Model Size=1B, Activat...
2025.05
41.23
Feedback
Search any
task
Search any
task