INDEX

Explanations

referring to people/groups

np_max-act · gemini-2.0-flash

The neuron activates on quotation marks and the words immediately surrounding quoted phrases.

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

Tour

-0.06

tx

-0.06

Eval

-0.06

Rub

-0.06

TS

-0.06

 intercepted

-0.06

मह

-0.06

 muss

-0.06

展示

-0.06

POSITIVE LOGITS

 zengin

0.08

 составе

0.06

اکم

0.06

-condition

0.06

ندگی

0.06

razione

0.06

 governo

0.06

 соблюд

0.06

 takové

0.06

 läng

0.06

Activations Density 0.090%

referring to people/groups

The neuron activates on quotation marks and the words immediately surrounding quoted phrases.

No Comments

No Known Activations