INDEX

Explanations

sentiment/language

np_max-act · gemini-2.0-flash

This neuron activates on numeric tokens, particularly decimal numbers (e.g. sentiment scores).

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

شتر

-0.07

perhaps

-0.07

 انر

-0.07

 Incontri

-0.06

qml

-0.06

 допомоги

-0.06

زه

-0.06

_survey

-0.06

Marc

-0.06

(curr

-0.06

POSITIVE LOGITS

0.07

pit

0.07

 gears

0.07

clr

0.06

 dış

0.06

 washer

0.06

erialization

0.06

DOWN

0.06

 Document

0.06

-status

0.06

Activations Density 0.001%

sentiment/language

This neuron activates on numeric tokens, particularly decimal numbers (e.g. sentiment scores).

No Comments

No Known Activations