INDEX

Explanations

numerical values, specifically those related to contact information and job listings

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

 calendriers

-0.77

ptonshire

-0.73

 ویکی‌پدیا

-0.73

uspiel

-0.70

 mergeFrom

-0.70

loride

-0.67

kloped

-0.66

 luck

-0.66

ulemon

-0.66

SharedDtor

-0.66

POSITIVE LOGITS

 chimiques

0.45

 charge

0.42

charge

0.41

 Charge

0.39

 chiam

0.38

lohy

0.38

 dochter

0.38

hile

0.38

parsedMessage

0.37

 Moms

0.36

Activations Density 0.037%

numerical values, specifically those related to contact information and job listings

No Comments

No Known Activations