INDEX

Explanations

step

np_max-act · gemini-2.0-flash

This neuron activates on the token “step,” as in the phrase “step-by-step” used in chain-of-thought reasoning.

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

 مسیر

-0.08

-holder

-0.07

 implementations

-0.07

 Desmond

-0.06

.TEXTURE

-0.06

Fol

-0.06

xfff

-0.06

Profiles

-0.06

 دست

-0.06

確認

-0.06

POSITIVE LOGITS

 vict

0.07

$instance

0.06

̀

0.06

شو

0.06

 Loki

0.06

á

0.06

 mennes

0.06

 tacos

0.06

 currentTime

0.06

Activations Density 0.005%

step

This neuron activates on the token “step,” as in the phrase “step-by-step” used in chain-of-thought reasoning.

No Comments

No Known Activations