INDEX

Explanations

capture

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

inant

-0.27

iant

-0.27

çĲĨè®º

-0.26

",__

-0.25

theory

-0.25

ctions

-0.25

 THEORY

-0.25

igy

-0.24

 theory

-0.24

Ã³g

-0.24

POSITIVE LOGITS

akens

0.31

reg

0.27

 ÑģÑĢÐ°Ð·Ñĥ

0.27

ä½ı

0.26

ä¿Ŀç®¡

0.26

mpr

0.26

åĶ®ç¥¨

0.24

åĽĽæĸ¹

0.24

 regs

0.24

odon

0.24

Activations Density 0.004%

capture

No Comments

No Known Activations