INDEX

Explanations

terms that reflect generalizations and broad categorizations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

出版年

-0.63

NameInMap

-0.57

olische

-0.57

oredCriteria

-0.56

yntaxException

-0.54

 vectorielle

-0.51

 susun

-0.50

 JAXBElement

-0.50

 ainfi

-0.50

 quæ

-0.50

POSITIVE LOGITS

 allgemein

0.58

 keseluruhan

0.57

 itself

0.56

ValueStyle

0.55

УДК

0.52

principalColumn

0.52

 broader

0.50

 wider

0.50

 khususnya

0.49

 generally

0.49

Activations Density 0.262%

terms that reflect generalizations and broad categorizations

No Comments

No Known Activations