INDEX

Explanations

sections of code related to styling and component structure in a programming context

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

technical code and foreign 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

TagMode

-0.88

transQ

-0.83

findpost

-0.77

 Meksiku

-0.73

Geplaatst

-0.71

 utafitiHapana

-0.71

jooq

-0.71

fromnode

-0.68

évaluateur

-0.67

تقاوى

-0.65

POSITIVE LOGITS

styleUrls

1.13

GeoNames

0.33

 yalnızca

0.32

 makeStyles

0.32

scss

0.32

AccessorTable

0.31

 estético

0.31

 connexe

0.30

 variantes

0.30

 connexes

0.30

Activations Density 0.000%

sections of code related to styling and component structure in a programming context

technical code and foreign words

No Comments

No Known Activations