INDEX

Explanations

references to photographs or imagery in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 Mumford

-0.55

 Modus

-0.52

 Kammer

-0.49

 appre

-0.48

主に

-0.48

 مظ

-0.48

 attributed

-0.47

 సౌకర్య

-0.47

Modus

-0.47

 contextLoads

-0.46

POSITIVE LOGITS

PHOTOS

0.81

HasAnnotation

0.80

Искәрмәләр

0.80

Photos

0.79

 تانيه

0.77

 Photos

0.76

 unknownFields

0.76

у

0.72

 Congrats

0.70

WireFormatLite

0.70

Activations Density 0.094%

references to photographs or imagery in text

No Comments

No Known Activations