INDEX

Explanations

total

np_max-act · gemini-2.0-flash

uses of the adjective denoting completeness/entirety, including capitalized occurrences as part of proper names or titles.

oai_token-act-pair · gpt-5 Triggered by @liuliujiujiu05

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_7.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.7.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aki

-0.29

resent

-0.28

op

-0.28

æ¬¢è¿İ

-0.27

iska

-0.27

 Sparks

-0.26

æľºä¼ļ

-0.26

å®Ŀ

-0.25

åıįæĺł

-0.25

å®ŀåĬĽ

-0.25

POSITIVE LOGITS

ONSE

0.31

UCE

0.29

è¿Ŀæ³ķè¿Ŀè§Ħ

0.28

biÃ³

0.28

alties

0.27

gzip

0.27

FML

0.26

å¤ªä¹ħ

0.25

 qualifies

0.25

æĪĳçŁ¥éģĵ

0.24

Activations Density 0.006%

total

uses of the adjective denoting completeness/entirety, including capitalized occurrences as part of proper names or titles.

No Comments

No Known Activations