INDEX

Explanations

references to programming classes and their functionalities

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

 informée

-0.68

chartInstance

-0.65

 незавершена

-0.64

rawDesc

-0.64

WireFormatLite

-0.64

oredCriteria

-0.63

addCriterion

-0.61

XtraReports

-0.60

onViewCreated

-0.59

 Савезне

-0.59

POSITIVE LOGITS

 contactez

0.43

 Aufwand

0.42

 tierras

0.42

 tirantes

0.41

 esclavos

0.41

 confirmer

0.40

 Moskau

0.40

 Endlich

0.40

 palvel

0.39

tainen

0.39

Activations Density 0.693%

references to programming classes and their functionalities

No Comments

No Known Activations