INDEX

Explanations

elements in a structured format, possibly related to data organization or programming

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

 FetchType

-0.49

DockStyle

-0.44

XtraBars

-0.42

rzost

-0.41

StructEnd

-0.40

AnimationsModule

-0.40

offsetof

-0.39

OrFail

-0.36

UseVisualStyle

-0.36

 amen

-0.36

POSITIVE LOGITS

 autorytatywna

0.89

Diweddarwch

0.52

Personensuche

0.52

scryfall

0.48

 GetEnumerator

0.45

 "{\"

0.45

 sanitarias

0.43

 null

0.41

null

0.41

Activations Density 1.035%

elements in a structured format, possibly related to data organization or programming

No Comments

No Known Activations