INDEX

Explanations

logical qualifiers and conditions in structured arguments or statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 Normdatei

-0.52

GoogleApiClient

-0.51

 banderas

-0.47

adaptiveStyles

-0.44

 فريبيس

-0.42

 botella

-0.42

oneofs

-0.41

Obrázky

-0.40

 EconPapers

-0.40

 firebaseConfig

-0.39

POSITIVE LOGITS

bandcamp

0.42

EEU

0.39

exitRule

0.39

ải

0.38

apatalk

0.38

 Abstra

0.37

 conforma

0.37

@[+][

0.37

 already

0.36

ItemGroup

0.36

Activations Density 0.101%

logical qualifiers and conditions in structured arguments or statements

No Comments

No Known Activations