INDEX

Explanations

structured scientific notation and mathematical symbols

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

detailed document metadata and computational/programming structures like indices, regions, and complex nested data formats.

oai_token-act-pair · claude-3-5-haiku-20241022 Triggered by @neilrathi

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

Prompts (Dashboard)

24,576 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

TagMode

-0.85

 surla

-0.81

 queſta

-0.77

featureID

-0.75

 ब्रेकडाउन

-0.75

 ویکی‌پدی

-0.73

 للاسماء

-0.73

RectangleBorder

-0.71

 cherchés

-0.70

ंदीखरीदारी

-0.69

POSITIVE LOGITS

is

0.45

0.43

for

0.42

0.41

 îles

0.41

Rel

0.41

0.40

to

0.40

Un

0.39

Activations Density 3.603%

structured scientific notation and mathematical symbols

detailed document metadata and computational/programming structures like indices, regions, and complex nested data formats.

No Comments

No Known Activations