INDEX

Explanations

aesthetic qualities

np_max-act · gemini-2.0-flash

adjectival judgments of appearance or quality, especially in “look/appear/seem” constructions (often comparative or evaluative descriptors).

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

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_11.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.11.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

loff

-0.28

ç«ĭä½ĵ

-0.26

 disap

-0.26

urga

-0.25

éĺħè¯»

-0.25

iosa

-0.25

usted

-0.25

èĵ¬

-0.24

#:

-0.24

Ð¸ÑĤÐµÐ»ÑĮÐ½Ð¾

-0.24

POSITIVE LOGITS

è½µ

0.30

è¯ı

0.29

æĶ¹æŃ£

0.25

 uyar

0.25

åĬŁ

0.25

lÃ¢

0.25

ä¸Ģå£°

0.25

 Hundred

0.24

é¢Ŀ

0.24

 StÃ¼ck

0.23

Activations Density 0.077%

aesthetic qualities

adjectival judgments of appearance or quality, especially in “look/appear/seem” constructions (often comparative or evaluative descriptors).

No Comments

No Known Activations