INDEX

Explanations

uncommon words including foreign words

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_24.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.24.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

å¯¹æłĩ

-0.28

eur

-0.28

ITH

-0.26

oyo

-0.26

äººå·¥

-0.25

æ¸ł

-0.24

 drive

-0.24

lete

-0.24

èĦ

-0.24

ä»¥ä¸º

-0.24

POSITIVE LOGITS

å£°éŁ³

0.28

èĪĴæľį

0.27

 emission

0.27

å£°åĵį

0.27

ctime

0.26

 uncomfortable

0.26

olumn

0.26

bol

0.26

 emitted

0.26

 sounded

0.25

Activations Density 0.895%

uncommon words including foreign words

No Known Activations