INDEX

Explanations

negative values or indications of decline or loss

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_10/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 increí

-0.66

Vidite

-0.64

 feroit

-0.63

 Majefty

-0.62

 שוליים

-0.61

 étoient

-0.60

 indígen

-0.60

 dieſer

-0.60

 ſich

-0.59

 avoient

-0.59

POSITIVE LOGITS

日至

0.40

–

0.40

…

0.39

 până

0.38

…

0.37

Tel

0.37

and

0.37

pushFollow

0.37

−

0.36

Activations Density 0.354%

negative values or indications of decline or loss

No Comments

No Known Activations