INDEX

Explanations

phrases related to user instructions and technical specifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_10/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

RectangleBorder

-0.49

ConstraintMaker

-0.45

transQ

-0.44

Muerte

-0.43

CheckBreak

-0.42

]^{-

-0.42

 illet

-0.41

 unknownFields

-0.40

devším

-0.40

kommun

-0.39

POSITIVE LOGITS

Derbyniad

0.58

elemField

0.47

GEBURTSDATUM

0.47

💼

0.45

tvguidetime

0.45

webElementXpaths

0.42

\{\\

0.42

postIndex

0.41

pot

0.41

Portail

0.41

Activations Density 0.259%

phrases related to user instructions and technical specifications

No Comments

No Known Activations