INDEX

Explanations

terms related to JSON editing and UI components

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

śnia

-0.55

FontOfSize

-0.52

Personendaten

-0.51

 AssemblyTitle

-0.50

+#+

-0.50

 saveiro

-0.48

KommentareTeilen

-0.47

 ladrillo

-0.47

endphp

-0.47

 opérations

-0.46

POSITIVE LOGITS

slidesToShow

0.41

 lenker

0.40

Autoritní

0.40

 pods

0.39

backends

0.39

 pist

0.37

 Италијани

0.37

 Wicidata

0.35

antd

0.35

ech

0.34

Activations Density 0.005%

terms related to JSON editing and UI components

No Comments

No Known Activations