INDEX

Explanations

COPD

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_1.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.1.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

åħ«å¤§

-0.28

æºĲæºĲä¸įæĸŃ

-0.26

èĮĥæĸĩ

-0.26

`[

-0.25

![

-0.25

steen

-0.25

æŁ³å·ŀ

-0.25

 comps

-0.24

Dam

-0.24

éĿ³

-0.24

POSITIVE LOGITS

 clinically

0.27

ä¾¿

0.27

Orig

0.26

ilon

0.26

è¾Ħ

0.25

urs

0.25

itaire

0.25

 Ã§Ä±k

0.25

urse

0.24

anja

0.24

Activations Density 12.476%

COPD

No Comments

No Known Activations