INDEX

Explanations

the

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

é¢Ī

-0.29

åıĭä»¬

-0.28

æĦı

-0.28

Injected

-0.26

é«ĺåľ°

-0.26

 Catal

-0.24

æľĢåħ³éĶ®çļĦ

-0.24

ÑģÑĤÐ²Ð°

-0.23

 Cree

-0.23

çļĦäººçī©

-0.23

POSITIVE LOGITS

thing

0.33

THING

0.31

 minute

0.30

thin

0.30

 thing

0.29

 hope

0.29

abouts

0.28

 corner

0.28

ä¸įåĲĮç¨ĭåº¦

0.28

 corresponding

0.28

Activations Density 0.039%

the

No Comments

No Known Activations