INDEX

Explanations

workers

np_max-act · gemini-2.0-flash

named-entity style tokens in running text, especially capitalized proper names, initials, and numeric figures.

oai_token-act-pair · gpt-5 Triggered by @liuliujiujiu05

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_11.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.11.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 susp

-0.32

con

-0.30

vac

-0.27

çĸĳ

-0.25

Ð°Ð½Ð¸Ñı

-0.25

til

-0.25

bian

-0.25

te

-0.25

 Ø§ÙĦÙħØ´

-0.24

 empty

-0.24

POSITIVE LOGITS

:last

0.27

.AttributeSet

0.27

Ip

0.26

Actor

0.25

 Breitbart

0.24

æĺ¯åĽłä¸º

0.24

ocrats

0.24

last

0.24

uters

0.24

 Chancellor

0.23

Activations Density 0.276%

workers

named-entity style tokens in running text, especially capitalized proper names, initials, and numeric figures.

No Comments

No Known Activations