INDEX

Explanations

formatted code elements or attributes

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

 Económica

-0.45

 Inscrivez

-0.44

 Nacionales

-0.41

 Könige

-0.41

 viajeros

-0.40

 alrededores

-0.39

 romántico

-0.38

 estatua

-0.38

 olvido

-0.38

ungkin

-0.37

POSITIVE LOGITS

 informée

0.77

!*\

0.75

 Infórmanos

0.70

fortawesome

0.66

saraba

0.59

PreferredItem

0.59

twimg

0.59

̈́

0.59

epam

0.58

تقاوى

0.58

Activations Density 1.232%

formatted code elements or attributes

No Comments

No Known Activations