INDEX

Explanations

phrases indicating relationships or connections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 Signalez

-0.63

 وتسجيلات

-0.59

kloped

-0.57

lete

-0.55

Bä

-0.53

 both

-0.52

Tembelea

-0.52

the

-0.51

istoitu

-0.50

ConstraintMaker

-0.49

POSITIVE LOGITS

eterangan

0.70

cèse

0.61

OrFail

0.61

 tartalomajánló

0.59

StateToProps

0.59

 kasarigan

0.59

stdarg

0.57

Lieferumfang

0.57

참고

0.55

asyarakat

0.55

Activations Density 0.304%

phrases indicating relationships or connections

No Comments

No Known Activations