INDEX

Explanations

no

np_max-act · gemini-2.0-flash

This neuron activates on numeric tokens, especially floating‐point numbers in response headers or logs.

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

achts

-0.06

�

-0.06

Sav

-0.06

 favourites

-0.06

tica

-0.06

MED

-0.06

_sale

-0.06

_sets

-0.06

 Aunt

-0.06

.answer

-0.06

POSITIVE LOGITS

oggles

0.07

 Petty

0.06

 преступ

0.06

Science

0.06

OMB

0.06

 orally

0.06

uteur

0.06

 přisp

0.06

xAF

0.06

(forKey

0.06

Activations Density 0.002%

no

This neuron activates on numeric tokens, especially floating‐point numbers in response headers or logs.

No Comments

No Known Activations

no

This neuron activates on numeric tokens, especially floating‐point numbers in response headers or logs.

No Comments

No Known Activations