INDEX

Explanations

all

np_max-act · gemini-2.0-flash

The neuron detects mentions of “all actions” being within the game’s (or simulation’s) boundaries—that is, scope-limiting disclaimers about actions.

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

 Marxism

-0.08

);

-0.06

Dro

-0.06

Ά

-0.06

 Σε

-0.06

Islamic

-0.06

realloc

-0.06

 Electro

-0.06

Reaction

-0.06

 практически

-0.06

POSITIVE LOGITS

 allowable

0.07

nell

0.07

 Asphalt

0.06

 đẹp

0.06

 extraordin

0.06

(fd

0.06

dependency

0.06

 cogn

0.06

dux

0.06

Activations Density 0.015%

all

The neuron detects mentions of “all actions” being within the game’s (or simulation’s) boundaries—that is, scope-limiting disclaimers about actions.

No Comments

No Known Activations