INDEX

Explanations

from

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

æ£ī

-0.30

ITCH

-0.28

 serialVersionUID

-0.28

ç¦Ģ

-0.27

ç¨ł

-0.26

ä¾ĿæĹ§

-0.26

itch

-0.26

çº±

-0.26

ä¾ĿçĦ¶

-0.26

<![

-0.25

POSITIVE LOGITS

 being

0.35

 becoming

0.32

 Blonde

0.28

çļĦè¦ģæ±Ĥ

0.27

ä¾Ĩèĩª

0.26

.getD

0.26

 reaching

0.26

being

0.26

å¯¹åħ¬åı¸

0.25

åģľäº§

0.25

Activations Density 0.005%

from

No Comments

No Known Activations