INDEX

Explanations

phrases related to measurement or relationship in contexts of compliance or combinations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

!*\

-0.76

:✨

-0.74

 &___

-0.74

afficheront

-0.73

 Autorizaciones

-0.72

WebElementEntity

-0.67

uxxxx

-0.66

HomeAsUpEnabled

-0.65

ItemBackground

-0.65

 '\\;'

-0.65

POSITIVE LOGITS

 combination

0.49

 full

0.49

 form

0.49

 harmony

0.48

zystać

0.47

 complete

0.47

 biais

0.46

 direct

0.46

 perfect

0.46

 close

0.45

Activations Density 0.388%

phrases related to measurement or relationship in contexts of compliance or combinations

No Comments

No Known Activations