INDEX

Explanations

names of authors, contributors, and publications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

fjspx

-0.74

enumi

-0.71

DoubleQuotes

-0.68

GIVEREF

-0.66

دانشنامهٔ

-0.65

unknownFields

-0.65

 propOrder

-0.65

withIdentifier

-0.63

INCREF

-0.63

GRANTED

-0.63

POSITIVE LOGITS

</table>

0.49

">//

0.44

ij

0.41

).\

0.40

 einzigen

0.39

tili

0.39

mane

0.39

manas

0.38

ophilus

0.38

timas

0.38

Activations Density 0.163%

names of authors, contributors, and publications

No Comments

No Known Activations