INDEX

Explanations

references to values and principles in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

й

-0.79

 DeWitt

-0.75

 Bess

-0.75

 Kongo

-0.73

 Margot

-0.72

 Mitar

-0.71

 Antiquities

-0.65

giy

-0.65

 Missy

-0.64

 Pfund

-0.64

POSITIVE LOGITS

 values

1.78

values

1.63

 Values

1.58

 VALUES

1.56

VALUES

1.47

Values

1.46

1.41

 Werte

1.10

 Valores

1.10

 valores

1.05

Activations Density 0.133%

references to values and principles in various contexts

No Comments

No Known Activations