Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Factoid Knowledge Recall on Key-Value Recall Factoid
Loading...
10.6
ID Score
REMIX (Random / K-Pile)
-0.32
2.515
5.35
8.185
Nov 11, 2024
ID Score
LAMA Score
EQ Score
WQ Score
Average Score
Updated 27d ago
Evaluation Results
Method
Method
Links
ID Score
LAMA Score
EQ Score
WQ Score
Average Score
REMIX (Random / K-Pile)
Backbone=Llama-3-8B, M...
2024.11
10.6
62.4
69.5
70.2
53.2
REMIX (Random / -)
Backbone=Llama-3-8B, M...
2024.11
8.9
2.5
42.5
61.4
28.8
REMIX (- / K-Pile)
Backbone=Llama-3-8B, M...
2024.11
0.8
40
36.4
33.9
27.8
No Mixing
Backbone=Llama-3-8B, M...
2024.11
0.5
2.1
17.4
33.8
13.5
REMIX (- / Random)
Backbone=Llama-3-8B, M...
2024.11
0.2
0.1
2.9
5.3
2.1
REMIX (K-Pile / -)
Backbone=Llama-3-8B, M...
2024.11
0.1
0
3.2
30.1
8.4
Feedback
Search any
task
Search any
task