INDEX

Explanations

noticing or placement

np_max-act · gemini-2.0-flash

mentions of the root “spot” in words, especially verb forms about noticing or identifying and names that include it.

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

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_1.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.1.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

eteria

-0.31

iquÃ©

-0.27

argas

-0.26

ä¸ĭçļĦ

-0.26

 frat

-0.25

 elementType

-0.25

éĵ

-0.25

ACS

-0.24

ECH

-0.24

 Kane

-0.23

POSITIVE LOGITS

izens

0.29

åĵģä½į

0.27

 deviation

0.25

åģıå·®

0.25

bef

0.24

 morph

0.24

çľŁç©º

0.24

 nominate

0.23

ise

0.23

onym

0.23

Activations Density 36.141%

noticing or placement

mentions of the root “spot” in words, especially verb forms about noticing or identifying and names that include it.

No Comments

No Known Activations