INDEX

Explanations

out

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

_su

-0.30

toFloat

-0.28

Occurrences

-0.27

ipsis

-0.27

 Authority

-0.26

-flash

-0.24

æŁĲä¸Ģ

-0.23

OPS

-0.23

æľ¯è¯Ń

-0.23

MarshalAs

-0.23

POSITIVE LOGITS

 there

0.35

0.31

 There

0.29

 then

0.29

There

0.28

 earlier

0.27

 otherwise

0.27

åĪĻ

0.27

etime

0.26

adj

0.26

Activations Density 0.017%

out

No Comments

No Known Activations