INDEX

Explanations

patterns related to numerical data or statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 19-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_19/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.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

 verband

-0.49

će

-0.45

 canne

-0.43

pair

-0.42

чет

-0.42

 engagé

-0.42

 excise

-0.42

ıştı

-0.42

 mottled

-0.42

ıştır

-0.41

POSITIVE LOGITS

 Савезне

0.84

 disambiguazione

0.80

хьтан

0.77

 سكانية

0.76

SourceChecksum

0.75

ніципалі

0.74

 المعيارى

0.73

UAWEI

0.72

 bezeichneter

0.68

 cherchés

0.67

Activations Density 0.398%

patterns related to numerical data or statistics

No Comments

No Known Activations