INDEX

Explanations

1

np_max-act · gemini-2.0-flash

The neuron selectively fires on translation‐file header metadata (e.g. fields like “MIME‐Version,” “X‐Crowdin…,” dates, version numbers, 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

 suç

-0.07

ISC

-0.07

ectors

-0.06

ctions

-0.06

adera

-0.06

Pager

-0.06

implified

-0.06

ิท

-0.06

omedical

-0.06

ifiers

-0.06

POSITIVE LOGITS

(IOException

0.07

 vaping

0.07

-leading

0.07

making

0.07

cat

0.07

 proportions

0.06

｀ヽ

0.06

ــ

0.06

 Chính

0.06

 bền

0.06

Activations Density 0.007%

1

The neuron selectively fires on translation‐file header metadata (e.g. fields like “MIME‐Version,” “X‐Crowdin…,” dates, version numbers, etc.).

No Comments

No Known Activations