INDEX

Explanations

changer

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

sun

-0.28

sun

-0.27

sure

-0.27

.sun

-0.26

 truly

-0.25

é¦

-0.25

 sinh

-0.25

ì¶ľ

-0.25

_sun

-0.25

js

-0.24

POSITIVE LOGITS

çİ·

0.27

æĶ¾å¤§

0.26

äººå¯¿

0.25

å¿ĥçĲĨ

0.25

nesia

0.24

çīį

0.24

å®Įå¥½

0.24

è¡¡

0.24

å¥ĳ

0.23

ä¿Ŀåģ¥

0.23

Activations Density 8.900%

changer

No Comments

No Known Activations