INDEX

Explanations

shape

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_14.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.14.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 Bernstein

-0.30

 facto

-0.28

éªĹåŃĲ

-0.28

éº¸

-0.27

åĺĪ

-0.26

ãģ£ãģ±

-0.25

(PR

-0.25

 statusBar

-0.24

unks

-0.24

iskey

-0.24

POSITIVE LOGITS

edere

0.33

 Ratings

0.32

 Rating

0.30

elle

0.28

 Finished

0.27

å®ĮäºĨ

0.25

éĩİ

0.25

 rating

0.25

 ratings

0.25

moid

0.25

Activations Density 0.185%

shape

No Known Activations