INDEX

Explanations

specific phrases indicating actions or suggestions

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

enlisted, SON, or rejected

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

AccessorTable

-0.70

 autorytatywna

-0.66

 ویکی‌پدی

-0.63

 Italijani

-0.60

 ब्रेकडाउन

-0.58

 AspNetCore

-0.57

ğraf

-0.57

 tartalomajánló

-0.54

WebElementEntity

-0.54

asteroido

-0.54

POSITIVE LOGITS

httphttps

0.50

findpost

0.37

 specie

0.36

ußer

0.35

 gall

0.35

 Anne

0.35

SON

0.35

 Pater

0.35

 Pete

0.34

jef

0.33

Activations Density 1.798%

specific phrases indicating actions or suggestions

enlisted, SON, or rejected

No Comments

No Known Activations