INDEX

Explanations

elements related to programming and code execution, specifically in the context of graphical manipulation and UI updates

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

setting or adding attributes

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

 صوتيه

-0.82

transQ

-0.66

 يتيمه

-0.66

tonode

-0.57

تقاوى

-0.57

Sklici

-0.52

 okuyayım

-0.48

 незавершена

-0.48

 betweenstory

-0.48

 queſta

-0.48

POSITIVE LOGITS

0.43

 Bonding

0.43

 söyle

0.42

ɵɵ

0.42

 inclined

0.41

 followed

0.40

滑

0.40

 ANNOUNC

0.40

 khoản

0.39

 Anmeldung

0.39

Activations Density 0.008%

elements related to programming and code execution, specifically in the context of graphical manipulation and UI updates

setting or adding attributes

No Comments

No Known Activations