INDEX

Explanations

gene-related information and annotations

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Sport

-0.37

 Power

-0.34

 Plan

-0.34

 Date

-0.34

 sold

-0.33

-0.32

Fo

-0.32

↵

-0.31

Mag

-0.31

POSITIVE LOGITS

 disambiguazione

0.85

 indígen

0.83

 estadounid

0.79

 increí

0.78

 desmotivaciones

0.73

 Italijani

0.73

rungsseite

0.72

 Bioinformatics

0.72

majánló

0.69

 plufieurs

0.69

Activations Density 1.960%

gene-related information and annotations

No Comments

No Known Activations