INDEX

Explanations

potential

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_27.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.27.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

ixmap

-0.26

èħ¹æ³»

-0.26

 Translation

-0.26

ellation

-0.26

Translation

-0.25

 pounded

-0.25

irl

-0.25

tridge

-0.25

ogh

-0.25

oples

-0.24

POSITIVE LOGITS

 ways

0.35

æ½ľåľ¨

0.32

 losses

0.27

 greatness

0.27

 potential

0.27

 Ways

0.26

 possibilities

0.26

ä¸įèī¯ä¿¡æģ¯

0.26

ureka

0.26

 similarities

0.26

Activations Density 0.006%

potential

No Comments

No Known Activations