INDEX

Explanations

phrases indicating requests for help or resources

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

wh

-0.58

-_-

-0.52

RTLR

-0.51

autique

-0.51

strike

-0.49

Hau

-0.47

Strike

-0.46

Rohy

-0.46

 strike

-0.46

PRIME

-0.46

POSITIVE LOGITS

<bos>

1.03

findpost

0.92

 مرئيه

0.72

 المعيارى

0.65

 متحده

0.65

ⓧ

0.65

 بيها

0.63

 createStore

0.61

########.

0.61

mybatisplus

0.60

Activations Density 0.064%

phrases indicating requests for help or resources

No Comments

No Known Activations