INDEX

Explanations

phrases and references that denote evidence or acknowledgment within a context

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

King Kong book writer

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

OGND

-0.77

 للمعارف

-0.74

 وتسجيلات

-0.71

 <=",

-0.70

 Normdatei

-0.68

حياته

-0.68

حياتها

-0.64

verwijspagina

-0.64

 Chwiliwch

-0.63

fjspx

-0.63

POSITIVE LOGITS

Release

0.38

 üzer

0.35

 release

0.34

 releases

0.32

 book

0.31

CommonModule

0.31

CollectionView

0.31

 originaux

0.31

 Release

0.30

 Sieger

0.30

Activations Density 0.012%

phrases and references that denote evidence or acknowledgment within a context

King Kong book writer

No Comments

No Known Activations