INDEX

Explanations

few

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_27.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.27.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

iminary

-0.26

 roadside

-0.25

iÄĻ

-0.24

wiki

-0.24

resher

-0.23

cul

-0.23

.rm

-0.22

 Ã¶l

-0.22

folk

-0.22

artial

-0.22

POSITIVE LOGITS

åıĳ

0.35

åĪļå¥½

0.28

 åıĳ

0.28

æŃ£å¥½

0.27

å¤´åıĳ

0.26

cene

0.26

OUNCE

0.26

éĩįå¤§

0.25

ÑĤÐµÐ½

0.25

frag

0.25

Activations Density 0.006%

few

No Comments

No Known Activations