INDEX

Explanations

]

np_max-act · gemini-2.0-flash

The neuron fires on the token “Instruction,” i.e. it detects when the instruction header appears.

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

Hog

-0.07

\Resources

-0.06

 Nigel

-0.06

_bid

-0.06

 Роб

-0.06

 Netanyahu

-0.06

 Consent

-0.06

noe

-0.06

네요

-0.06

 SpaceX

-0.06

POSITIVE LOGITS

_PROVIDER

0.07

vais

0.07

編

0.06

ponsored

0.06

реди

0.06

locale

0.06

/');↵

0.06

__(

0.06

setChecked

0.06

 offsetY

0.06

Activations Density 0.003%

]

The neuron fires on the token “Instruction,” i.e. it detects when the instruction header appears.

No Comments

No Known Activations