INDEX

Explanations

numbers

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

scenes

-0.28

 scenes

-0.26

èĬľ

-0.25

éĻĲãĤĬ

-0.24

-scenes

-0.24

getLocation

-0.24

é¦Ĵ

-0.24

 diss

-0.24

mdir

-0.24

 Leicester

-0.23

POSITIVE LOGITS

vod

0.31

å·²æľī

0.27

åİļ

0.26

ÙģØ©

0.26

vo

0.26

ashing

0.25

BJ

0.25

å¤§éĩı

0.25

è¦ģç´ł

0.25

éļĻ

0.24

Activations Density 0.006%

numbers

No Comments

No Known Activations