INDEX

Explanations

theorems and results in a formal mathematical context

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

(['/

-0.43

gypti

-0.43

eqref

-0.42

 souci

-0.41

rowse

-0.41

NewLine

-0.40

 MÉXICO

-0.40

越

-0.40

 universe

-0.40

":["

-0.39

POSITIVE LOGITS

 StatelessWidget

0.65

 ComVisible

0.64

Corollary

0.56

URLException

0.53

 AssemblyCulture

0.52

 StatefulWidget

0.51

ElementException

0.51

 HttpStatusCode

0.49

ArgsConstructor

0.45

 XCTest

0.45

Activations Density 0.962%

theorems and results in a formal mathematical context

No Comments

No Known Activations