INDEX

Explanations

code and math

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_18.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.18.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Kann

-0.26

å¿Į

-0.25

le

-0.25

æĺİäº®

-0.25

éĢģ

-0.25

 snaps

-0.25

ymax

-0.24

 climbing

-0.24

çĽĺ

-0.24

 implants

-0.24

POSITIVE LOGITS

æĬ¥è®°èĢħ

0.28

è¦ı

0.27

future

0.26

ç»ħ

0.25

 Furn

0.25

æ²Ļåıĳ

0.25

 patented

0.24

åįģåħ«å±Ĭ

0.24

 Tribune

0.24

 future

0.24

Activations Density 14.359%

code and math

No Comments

No Known Activations