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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

quet

-0.25

.aw

-0.24

iller

-0.24

rah

-0.24

oss

-0.24

Ã¼nst

-0.24

 cÃ²n

-0.23

åĿĩæľī

-0.23

éĻĲé¢Ŀ

-0.23

AW

-0.23

POSITIVE LOGITS

verity

0.29

ç¼ł

0.28

InitialState

0.27

åĪĿæ¬¡

0.27

TickCount

0.27

 Suff

0.26

sville

0.26

mare

0.25

/sources

0.25

ppo

0.25

Activations Density 0.026%

numbers

No Comments

No Known Activations