INDEX

Explanations

elements related to customizing digital interfaces and personalization features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

WriteBarrier

-0.67

 intStringLen

-0.54

lovakia

-0.50

SharedDtor

-0.50

RetentionPolicy

-0.48

 continúas

-0.47

 "..\..\..\

-0.46

RUnlock

-0.43

 queſta

-0.43

]=-

-0.42

POSITIVE LOGITS

 giggles

0.38

 selfie

0.38

selfie

0.38

+:+

0.37

 selfies

0.37

PSD

0.37

Make

0.36

 Selfie

0.36

ajuku

0.36

CAG

0.35

Activations Density 0.007%

elements related to customizing digital interfaces and personalization features

No Comments

No Known Activations