INDEX

Explanations

patterns related to data structure elements and their relationships

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

delimiter followed by angle bracket

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

 queſta

-0.82

IntoConstraints

-0.76

 ModelExpression

-0.73

 برانيه

-0.67

 ſind

-0.66

NSIndexPath

-0.65

$_"

-0.64

iſten

-0.64

 awakeFromNib

-0.64

 kasarigan

-0.64

POSITIVE LOGITS

“

0.56

 ponerse

0.54

 either

0.54

 světě

0.49

 libremente

0.49

0.48

 presentarse

0.48

 říká

0.47

 Geheimnis

0.47

 gewöhnlich

0.47

Activations Density 0.104%

patterns related to data structure elements and their relationships

delimiter followed by angle bracket

No Comments

No Known Activations