INDEX

Explanations

%

np_max-act · gemini-2.0-flash

This neuron activates on LLVM IR SSA control-flow constructs—specifically lines with branch predecessor lists, “phi” nodes, and related metadata.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

оч

-0.07

vrd

-0.06

.CONTENT

-0.06

OX

-0.06

’’

-0.06

ắn

-0.06

 lords

-0.06

&,

-0.06

Showing

-0.06

ufig

-0.06

POSITIVE LOGITS

 preds

0.06

 enhancements

0.06

 clot

0.06

tot

0.06

 bakım

0.06

 eğit

0.06

 coolant

0.06

icans

0.06

rase

0.06

NOI

0.06

Activations Density 0.001%

%

This neuron activates on LLVM IR SSA control-flow constructs—specifically lines with branch predecessor lists, “phi” nodes, and related metadata.

No Comments

No Known Activations