INDEX

Explanations

Coding/technical context

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_12.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.12.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

::$

-0.29

ç®Ĺ

-0.28

åĬ«

-0.27

(ext

-0.25

å°ģ

-0.25

 Speedway

-0.24

 gent

-0.24

æĭīæĸ¯

-0.24

æģĲæĢĸ

-0.24

 GENERIC

-0.24

POSITIVE LOGITS

åľ¨è¿ĩæ¸¡

0.32

Py

0.29

çĹ¢

0.28

åĮ¾

0.28

Py

0.28

 KÃ¶nig

0.27

 Truth

0.26

è¿ĩæ¸¡

0.26

py

0.25

Falsy

0.25

Activations Density 0.001%

Coding/technical context

No Known Activations