INDEX

Explanations

phrases related to coding or technical instructions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/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.34.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

########.

-0.35

 derajat

-0.33

)>>

-0.33

KURZBESCHREIBUNG

-0.32

Diámetro

-0.32

ỡng

-0.31

 suelo

-0.31

IntoConstraints

-0.30

mergeFrom

-0.30

Concerning

-0.30

POSITIVE LOGITS

 Malaysian

0.72

 Kuala

0.70

 Singapore

0.69

BASEPATH

0.69

 Malay

0.66

 Malaysia

0.65

Singapore

0.63

CI

0.63

 Penang

0.58

 Borneo

0.58

Activations Density 0.037%

phrases related to coding or technical instructions

No Comments

No Known Activations