INDEX

Explanations

I

np_max-act · gemini-2.0-flash

This neuron activates on the assistant’s self-introduction phrase (e.g. “My name is Claude”).

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

emperature

-0.07

ERC

-0.07

 BIOS

-0.07

RAD

-0.07

NOT

-0.07

 Contrast

-0.06

 kidding

-0.06

前に

-0.06

Limited

-0.06

America

-0.06

POSITIVE LOGITS

 woodland

0.07

ヴ

0.06

(Set

0.06

◆

0.06

 etki

0.06

(gr

0.06

 तरह

0.06

，而且

0.06

 xứ

0.06

_CODES

0.06

Activations Density 0.031%

I

This neuron activates on the assistant’s self-introduction phrase (e.g. “My name is Claude”).

No Comments

No Known Activations

I

This neuron activates on the assistant’s self-introduction phrase (e.g. “My name is Claude”).

No Comments

No Known Activations