INDEX

Explanations

references to popular or notable literature and media

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Tokens after the word "is"

np_acts-logits-general · gemini-2.0-flash

is a designation

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 OMITBAD

-0.48

новниш

-0.47

 tantum

-0.40

省市镇

-0.40

 černá

-0.39

 następujące

-0.35

IsContent

-0.35

DoubleQuotes

-0.35

 nahilalakip

-0.35

cshtml

-0.34

POSITIVE LOGITS

AndEndTag

0.48

 champion

0.48

 препратки

0.47

ьаж

0.47

styleType

0.44

ConstraintMaker

0.43

 فريبيس

0.42

utives

0.42

enumii

0.42

 phố

0.41

Activations Density 0.143%

references to popular or notable literature and media

Tokens after the word "is"

is a designation

No Comments

No Known Activations