INDEX

Explanations

blog posts and introductions

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bsub

-0.28

ä»ħæ¬¡äºİ

-0.28

RIA

-0.27

Gra

-0.26

straction

-0.25

 granny

-0.25

Gn

-0.25

 juni

-0.25

ixa

-0.25

Wit

-0.24

POSITIVE LOGITS

 thereafter

0.50

 thereof

0.50

 afterward

0.46

ä¸ºæŃ¤

0.46

 afterwards

0.44

å¯¹æŃ¤

0.41

 accordingly

0.39

äºİæŃ¤

0.37

 ones

0.36

 meanwhile

0.34

Activations Density 0.004%

blog posts and introductions

No Comments

No Known Activations