INDEX

Explanations

patterns related to negative values or ratings

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

MigrationBuilder

-0.79

 تضيفلها

-0.75

:✨

-0.74

دانشنامهٔ

-0.65

SequentialGroup

-0.65

Personensuche

-0.64

SourceChecksum

-0.64

protoimpl

-0.64

saraba

-0.63

 GenerationType

-0.62

POSITIVE LOGITS

0.49

Dichloropropane

0.39

↵↵

0.37

↵

0.34

 abstract

0.33

0.32

0.31

Activations Density 0.112%

patterns related to negative values or ratings

No Comments

No Known Activations