INDEX

Explanations

etyl

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

 phantom

-0.30

 Phantom

-0.30

 panic

-0.28

omens

-0.27

rium

-0.26

ç»´å¥ĩ

-0.26

 anthem

-0.25

æĦıèŃĺ

-0.25

 Îº

-0.25

panic

-0.25

POSITIVE LOGITS

 automation

0.30

automation

0.29

 ofType

0.27

èĩªåĬ¨åĮĸ

0.26

cean

0.24

 subtract

0.24

gere

0.24

iners

0.24

qtt

0.24

 numbered

0.24

Activations Density 22.271%

etyl

No Comments

No Known Activations