INDEX

Explanations

references to data processing, invention, and technical specifications in various contexts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 zoude

-0.50

ſchaft

-0.50

 laſſen

-0.49

 nothwendig

-0.49

 sasuke

-0.47

vacc

-0.47

 surla

-0.46

 depic

-0.46

征詢我

-0.45

hoeddwyd

-0.44

POSITIVE LOGITS

GEBURTSDATUM

0.44

Personendaten

0.44

 autorytatywna

0.40

ValueGenerated

0.40

0.38

 AssemblyTitle

0.38

 תש

0.36

 CreateTagHelper

0.36

Mil

0.36

 estre

0.35

Activations Density 0.044%

references to data processing, invention, and technical specifications in various contexts

No Known Activations