INDEX

Explanations

with

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_7.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.7.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

.IsValid

-0.28

è·Łå¥¹

-0.27

è£Ļ

-0.26

èģĺçĶ¨

-0.25

 Portug

-0.25

åįĬä¸ª

-0.25

ä¸įéĢļ

-0.25

 wearing

-0.25

äº¨

-0.24

åİ»åģļ

-0.24

POSITIVE LOGITS

 seemed

0.26

allel

0.26

-lat

0.26

ero

0.25

lum

0.25

 seems

0.24

å¯¼

0.24

[];↵

0.24

Amazon

0.24

 SetProperty

0.24

Activations Density 0.002%

with

No Comments

No Known Activations