INDEX

Explanations

keywords related to animation effects in programming or user interface contexts

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

animate, animated, active

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

URLException

-0.84

transQ

-0.72

 endforeach

-0.70

StoryboardSegue

-0.69

󠁿

-0.69

 FetchType

-0.68

majánló

-0.67

 miniaturka

-0.66

 queſta

-0.65

Життєпис

-0.65

POSITIVE LOGITS

animate

1.12

 animate

0.89

Animate

0.79

 Animate

0.68

 animated

0.52

Animated

0.51

animated

0.47

 Animated

0.44

 anima

0.35

 anim

0.34

Activations Density 0.003%

keywords related to animation effects in programming or user interface contexts

animate, animated, active

No Comments

No Known Activations