INDEX

Explanations

Technical writing snippets

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

Negative Logits

own

-0.26

————————————————

-0.25

altern

-0.24

vg

-0.24

subst

-0.23

 Castillo

-0.23

.present

-0.22

Track

-0.22

inf

-0.22

own

-0.22

POSITIVE LOGITS

æ¯Ķå¦Ĥ

0.82

ä¾ĭå¦Ĥ

0.70

æ¯Ķå¦Ĥè¯´

0.69

è¯¸å¦Ĥ

0.59

å¦Ĥ

0.54

èŃ¬å¦Ĥ

0.54

ä¾ĭãģĪãģ°

0.53

 ÐĿÐ°Ð¿ÑĢÐ¸Ð¼ÐµÑĢ

0.47

Ð½Ð°Ð¿ÑĢÐ¸Ð¼ÐµÑĢ

0.44

 like

0.43

Activations Density 0.008%

Technical writing snippets

No Known Activations