Datasets:

MInAlA
/

crosscoder-multilayer-split-activations

Tags:

mechanistic-interpretability

sparse-autoencoder

License:

Dataset card Files Files and versions

crosscoder-multilayer-split-activations / v2-differentlayer /qwen3-4b-ppo /singlelayer_reduce /L20-22 /plots

3.27 MB

1 contributor

History: 1 commit

ishangarg183's picture

Copy assembled_activations to v2-differentlayer (batch 7/12)

282d256 verified about 1 month ago

cf_distribution_per_class.png

169 kB
xet

Copy assembled_activations to v2-differentlayer (batch 7/12) about 1 month ago
cf_shift_per_class.png

124 kB
xet

Copy assembled_activations to v2-differentlayer (batch 7/12) about 1 month ago
class_distribution.png

156 kB
xet

Copy assembled_activations to v2-differentlayer (batch 7/12) about 1 month ago
loss_curves.png

303 kB
xet

Copy assembled_activations to v2-differentlayer (batch 7/12) about 1 month ago
rho_histogram.png

82.5 kB
xet

Copy assembled_activations to v2-differentlayer (batch 7/12) about 1 month ago
rho_theta_scatter.png

490 kB
xet

Copy assembled_activations to v2-differentlayer (batch 7/12) about 1 month ago
shared_geometry_histograms.png

147 kB
xet

Copy assembled_activations to v2-differentlayer (batch 7/12) about 1 month ago
shared_geometry_theta_vs_norm_ratio.png

702 kB
xet

Copy assembled_activations to v2-differentlayer (batch 7/12) about 1 month ago
superposition_analysis.png

1.09 MB
xet

Copy assembled_activations to v2-differentlayer (batch 7/12) about 1 month ago