INDEX

Explanations

speaking/phone

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

Not in Any Lists

No Comments

Negative Logits

ç¦§

-0.27

 fertile

-0.25

_ve

-0.25

º«

-0.25

stra

-0.24

ception

-0.24

nun

-0.24

lap

-0.24

 Shed

-0.23

éĩĩ

-0.23

POSITIVE LOGITS

afs

0.31

enty

0.29

éļĥ

0.28

kest

0.27

crest

0.27

èµĭäºĪ

0.27

pery

0.26

uly

0.26

-notch

0.26

 Ð¾ÑĤÐ´

0.25

Activations Density 1.033%

speaking/phone

No Comments

No Known Activations