INDEX

Explanations

references to court cases and legal terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

entyfik

-0.36

 pilotes

-0.32

 nonatomic

-0.32

ęd

-0.31

 Konsequ

-0.31

stęp

-0.30

 vérit

-0.28

ワイイ

-0.28

 tape

-0.28

ṳ

-0.27

POSITIVE LOGITS

OGND

0.79

XmlAccessType

0.77

InitVars

0.77

ulemon

0.77

Vidite

0.73

ammans

0.71

SharedCtor

0.69

ThroughAttribute

0.69

 defaultstate

0.69

 XHTML

0.68

Activations Density 1.917%

references to court cases and legal terminology

No Comments

No Known Activations