INDEX

Explanations

phrases indicating conflict or threats

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

 وتسجيلات

-0.42

 Ries

-0.39

Linki

-0.39

 مشين

-0.39

 monotony

-0.39

Bibliografia

-0.38

 Amalgam

-0.38

참고

-0.38

waitKey

-0.37

 oppervlak

-0.37

POSITIVE LOGITS

DockStyle

0.64

Cyfarwyddwr

0.49

 juſ

0.48

 sanctions

0.47

 consecuencias

0.44

 dieß

0.43

 viſ

0.42

 consequ

0.42

IVEREF

0.42

 sanction

0.42

Activations Density 0.441%

phrases indicating conflict or threats

No Comments

No Known Activations