INDEX

Explanations

negations and phrases indicating an absence or lack of something

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

the followed by capitalized words

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 հղումներ

-0.45

bro

-0.44

formik

-0.43

sukienka

-0.43

ništ

-0.43

Étymologie

-0.43

Bioaccumulative

-0.41

 colorWith

-0.41

 Haller

-0.41

出版年

-0.41

POSITIVE LOGITS

expandindo

0.51

0.40

 defaultstate

0.39

 CreateTagHelper

0.37

XmlAccessorType

0.36

IUrlHelper

0.36

atchewan

0.35

OGND

0.34

roek

0.34

migrationBuilder

0.33

Activations Density 0.029%

negations and phrases indicating an absence or lack of something

the followed by capitalized words

No Comments

No Known Activations