INDEX

Explanations

repeat

np_max-act · gemini-2.0-flash

The neuron detects user requests phrased as “please repeat” or similar commands to repeat or echo text.

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

 integrate

-0.06

 seat

-0.06

 calculate

-0.06

 Significant

-0.06

mount

-0.06

 computed

-0.06

posit

-0.06

 Developing

-0.06

 Courage

-0.06

 seats

-0.06

POSITIVE LOGITS

_hidden

0.06

arbeit

0.06

 hors

0.06

üç

0.06

()){

0.06

 playback

0.06

탄

0.06

Nay

0.06

(varargin

0.06

PMC

0.06

Activations Density 0.018%

repeat

The neuron detects user requests phrased as “please repeat” or similar commands to repeat or echo text.

No Comments

No Known Activations