Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Fine-tuning on FineWeb (test)
Loading...
1.93
Test Loss
SISA
1.7992
2.6821
3.565
4.4479
Feb 15, 2025
Test Loss
Semantic Similarity
Updated 18d ago
Evaluation Results
Method
Method
Links
Test Loss
Semantic Similarity
SISA
Model=GPT2(335M), Epoc...
2025.02
1.93
84.84
AdamW
Model=GPT2(335M), Epoc...
2025.02
1.95
84.19
SISA
Model=GPT2(335M), Epoch=5
2025.02
2.04
85.24
SISA
Model=GPT2(124M), Epoc...
2025.02
2.17
84.82
AdamW
Model=GPT2(124M), Epoc...
2025.02
2.3
84.73
SISA
Model=GPT2(335M), Epoch=2
2025.02
2.31
84.76
SISA
Model=GPT2(124M), Epoch=5
2025.02
2.73
84.36
AdamW
Model=GPT2(335M), Epoch=5
2025.02
2.73
84.5
AdamW
Model=GPT2(124M), Epoch=5
2025.02
3.45
84.3
AdamW
Model=GPT2(335M), Epoch=2
2025.02
3.62
83.17
SISA
Model=GPT2(124M), Epoch=2
2025.02
3.69
83.15
AdamW
Model=GPT2(124M), Epoch=2
2025.02
5.2
82.16
Feedback
Search any
task
Search any
task
