Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Llama-3.2-1B

Benchmarks

Task NameDataset NameSOTA ResultTrend
Sparse AutoencodingLlama-3.2-1B Layer 12 activations (held-out)
Relative Error0.499
4
Showing 1 of 1 rows