INDEX

Explanations

I

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

è¯´å¾Ĺ

-0.28

çħ§æł·

-0.28

avar

-0.28

ç¢§

-0.26

èī²æ³½

-0.26

 Saying

-0.24

ÐµÑģÑĤÐ²Ð¾

-0.24

 Rear

-0.24

 Blast

-0.23

Ð´ÑĢ

-0.23

POSITIVE LOGITS

|null

0.30

æĶ¯

0.26

ohan

0.26

 plant

0.25

WAYS

0.25

jie

0.24

éĻĶ

0.24

pcb

0.24

edom

0.24

é¥Ń

0.24

Activations Density 0.001%

I

No Comments

No Known Activations