INDEX

Explanations

different

np_max-act · gemini-2.0-flash

This neuron detects action-oriented instructional phrases—especially “Experiment with different …” style commands in step-by-step advice.

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

-negative

-0.07

heed

-0.06

,U

-0.06

.he

-0.06

 Feld

-0.06

Speaker

-0.06

 Marcus

-0.06

Pal

-0.06

 PREFIX

-0.06

 lookup

-0.06

POSITIVE LOGITS

_music

0.07

uits

0.07

_TRAN

0.06

appro

0.06

 ])↵

0.06

�

0.06

 recept

0.06

_Desc

0.06

 एप

0.06

 vibr

0.06

Activations Density 0.022%

different

This neuron detects action-oriented instructional phrases—especially “Experiment with different …” style commands in step-by-step advice.

No Comments

No Known Activations