INDEX

Explanations

references to theorems, lemmas, propositions, and equations in mathematical text

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

 ddelweddau

-0.51

PointerException

-0.47

 resourceCulture

-0.46

EndTag

-0.46

 ligiloj

-0.43

 propOrder

-0.43

󠁮

-0.43

HtmlAttribute

-0.42

Diwedd

-0.41

ymce

-0.41

POSITIVE LOGITS

 Appendix

0.45

Appendix

0.44

prefix

0.39

Chapter

0.39

ADORA

0.37

IV

0.37

Dur

0.37

OUG

0.37

IVO

0.36

 Doug

0.36

Activations Density 1.032%

references to theorems, lemmas, propositions, and equations in mathematical text

No Comments

No Known Activations