INDEX

Explanations

open

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_33.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.33.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

They

-0.44

ä½Ĩä»ĸä»¬

-0.36

ä»ĸä»¬æĺ¯

-0.35

they

-0.35

Their

-0.33

ä»ĸä»¬

-0.33

 they

-0.33

 They

-0.32

ä»ĸä»¬åľ¨

-0.32

ffffffff

-0.31

POSITIVE LOGITS

UIL

0.28

upo

0.28

()>

0.27

up

0.26

/import

0.26

furt

0.26

/disc

0.25

/register

0.25

PRESS

0.25

 accordingly

0.25

Activations Density 0.033%

open

No Comments

No Known Activations