INDEX

Explanations

I

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

åĸ½

-0.29

åĻ«

-0.29

ä¸įæĪĲ

-0.27

çļĦçģµéŃĤ

-0.26

linger

-0.26

æĹº

-0.25

['$

-0.25

owie

-0.25

åķ¦

-0.24

æ²¡éĶĻ

-0.24

POSITIVE LOGITS

said

0.35

 said

0.33

æīĢåľ¨

0.30

 Said

0.29

edly

0.27

èĤ¢

0.26

æīĢè¿°

0.26

 mentioned

0.25

RB

0.25

æĭ£

0.25

Activations Density 0.001%

I

No Comments

No Known Activations