INDEX

Explanations

file paths

np_max-act · gemini-2.0-flash

The neuron activates on tokens that are parts of file or namespace import paths (e.g. directory‐separated identifiers and slashes/backslashes).

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

 Mods

-0.07

 boarded

-0.06

 walnut

-0.06

Ike

-0.06

 moden

-0.06

شی

-0.06

 deben

-0.06

_plate

-0.06

OfMonth

-0.06

 postponed

-0.06

POSITIVE LOGITS

footer

0.07

_top

0.07

�

0.06

fd

0.06

(`

0.06

/bg

0.06

쉬

0.06

初始化

0.06

Snackbar

0.06

ease

0.06

Activations Density 0.025%

file paths

The neuron activates on tokens that are parts of file or namespace import paths (e.g. directory‐separated identifiers and slashes/backslashes).

No Comments

No Known Activations