INDEX

Explanations

numeric values and their corresponding tags or labels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

 tdci

-0.75



-0.74

wpi

-0.73

 crdi

-0.69

AxisAlignment

-0.67

NewUrlParser

-0.66

Hozzáférés

-0.66

asteroide

-0.66

 Geraadpleegd

-0.66

 Elvira

-0.65

POSITIVE LOGITS

Carriera

0.43

 jail

0.42

lui

0.41

OPHER

0.40

 civiles

0.40

 none

0.40

 agree

0.40

 нет

0.39

דינה

0.39

 traseira

0.38

Activations Density 0.040%

numeric values and their corresponding tags or labels

No Comments

No Known Activations