INDEX

Explanations

occurrences of numeric values

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

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

PMailer

-0.47

 defaultstate

-0.47

EndContext

-0.46

出版年

-0.41

UnsafeEnabled

-0.41

 Numerade

-0.41

RTRS

-0.40

Források

-0.37

MultipartFile

-0.37

Bakgrunnsstoff

-0.37

POSITIVE LOGITS

🟤

0.53

 NSCoder

0.48

bibfield

0.48

legd

0.47

arriver

0.45

endpush

0.44

ConfigureAwait

0.44

 wireType

0.44

abatic

0.43

chevêque

0.42

Activations Density 0.650%

occurrences of numeric values

No Comments

No Known Activations