INDEX

Explanations

(

np_max-act · gemini-2.0-flash

This neuron fires on the marked‐answer option in multiple‐choice questions, i.e. the parenthesized choice label (a, b, c, etc.) together with its associated value.

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

Quant

-0.07

пор

-0.07

 moth

-0.06

_dict

-0.06

แล

-0.06

Word

-0.06

Descri

-0.06

 partes

-0.06

Cette

-0.06

ursos

-0.06

POSITIVE LOGITS

ezier

0.08

	admin

0.07

 sunscreen

0.06

-G

0.06

panies

0.06

 danced

0.06

ischen

0.06

 softened

0.06

міністра

0.06

HA

0.06

Activations Density 0.003%

(

This neuron fires on the marked‐answer option in multiple‐choice questions, i.e. the parenthesized choice label (a, b, c, etc.) together with its associated value.

No Comments

No Known Activations