INDEX

Explanations

ic

np_max-act · gemini-2.0-flash

This neuron detects words ending in the suffix “-ic,” such as chemical adjectives (e.g. “ferric,” “sulfuric,” etc.).

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

 actu

-0.07

افته

-0.07

 Dynamo

-0.07

.substr

-0.06

.prevent

-0.06

ystery

-0.06

 verde

-0.06

idders

-0.06

랍니다

-0.06

 assembly

-0.06

POSITIVE LOGITS

_End

0.08

 lastIndex

0.07

总

0.06

ων

0.06

ON

0.06

AGES

0.06

ritel

0.06

Can

0.06

 taco

0.06

.can

0.06

Activations Density 0.004%

ic

This neuron detects words ending in the suffix “-ic,” such as chemical adjectives (e.g. “ferric,” “sulfuric,” etc.).

No Comments

No Known Activations