INDEX

Explanations

world

np_max-act · gemini-2.0-flash

The neuron detects occurrences of the word stem “real” (as in “real” or “reality,” including inflected forms like Italian “realtà”).

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

 healed

-0.06

δό

-0.06

_lost

-0.06

zig

-0.06

terminate

-0.06

lık

-0.06

ró

-0.06

ώ

-0.06

SP

-0.06

eum

-0.06

POSITIVE LOGITS

 RFID

0.07

Rotor

0.06

 tekrar

0.06

 případě

0.06

 признач

0.06

 AGRE

0.06

Col

0.06

.Native

0.06

 polov

0.06

 CENTER

0.06

Activations Density 0.029%

world

The neuron detects occurrences of the word stem “real” (as in “real” or “reality,” including inflected forms like Italian “realtà”).

No Comments

No Known Activations