INDEX

Explanations

,

np_max-act · gemini-2.0-flash

This neuron activates primarily on document-structure tokens (e.g. <|start_header_id|>, <|eot_id|>, header/footer markers) rather than ordinary words.

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

($)

-0.07

 Prote

-0.07

 Prescott

-0.06

 slander

-0.06

don

-0.06

!=↵

-0.06

 MESSAGE

-0.06

 tertiary

-0.06

Ded

-0.06

(pref

-0.06

POSITIVE LOGITS

 abol

0.07

 समय

0.06

 identifiable

0.06

 Spacer

0.06

 sqlSession

0.06

 buscar

0.06

ys

0.06

hem

0.06

ือข

0.06

 ánh

0.06

Activations Density 0.107%

,

This neuron activates primarily on document-structure tokens (e.g. <|start_header_id|>, <|eot_id|>, header/footer markers) rather than ordinary words.

No Comments

No Known Activations