framework
active
framework:sparsely-connected-cross-layer-transcodersSparsely connected cross-layer transcoders
Mask virtual weights during training targeting interference.
Neighborhood — ranked by edge-count
Papers (1)
paper
Frameworks (1)
framework
- Cross-Layer Transcodersrelated_toMLP replacement with sparse features underlying the original attribution graphs.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Used to express MLP in monosemantic features, 4096 features with JumpReLU.
- Replacement model architecture with 30M features replacing MLP neurons.
- Extension of SAEs that jointly learns latents across representations from different models; proposed as alternative for extended fine-tuning
- Discussion.
- Sparse dictionary learning variant that reads and writes to multiple layers or models, yielding shared features.
- Uses activations from one layer to predict the next; special case of crosscoders.
- Learn a shared feature space across two models; basis of the work.
- Decomposition method for activations; VPD is compared against transcoders in sparsity-reconstruction tradeoff.