INDEX

Explanations

references to various exhibitions and their features

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

 conta

-0.74

 Pome

-0.73

pung

-0.72

</b>

-0.69

 оригіналу

-0.67

Www

-0.67

-0.66

seq

-0.64

manan

-0.64

kro

-0.63

POSITIVE LOGITS

 Exhibition

1.59

 EXHIB

1.57

 exhibitions

1.54

 exhibition

1.50

 Exhibitions

1.49

 exhibits

1.47

Exhibition

1.46

 Exhibits

1.42

 Exhibit

1.42

 exhibit

1.42

Activations Density 0.114%

references to various exhibitions and their features

No Comments

No Known Activations