INDEX

Explanations

keywords related to reports and investigations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

 taught

-0.37

probieren

-0.32

 teaches

-0.32

 erfahren

-0.30

 jokes

-0.30

regex

-0.29

hrungen

-0.28

 superstitious

-0.28

 Geschichten

-0.28

 samedi

-0.28

POSITIVE LOGITS

 report

1.73

 Report

1.58

Report

1.41

report

1.40

 REPORT

1.27

报告

1.26

 relatório

1.17

報告

1.16

 raport

1.16

REPORT

1.13

Activations Density 0.578%

keywords related to reports and investigations

No Comments

No Known Activations