INDEX

Explanations

references to places and associated settings or locations

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

the X

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 informée

-0.54

numerusform

-0.48

 跳转至

-0.45

plato

-0.43

styleable

-0.41

consulté

-0.41

 phosphates

-0.40

henge

-0.40

 obč

-0.39

 урна

-0.39

POSITIVE LOGITS

 Италијани

0.44

setVerticalGroup

0.44

 froide

0.40

 ویکی‌پدی

0.40

Démographie

0.39

Handlung

0.38

 oprot

0.37

sorry

0.36

 firebaseConfig

0.35

 kohdetta

0.34

Activations Density 0.034%

references to places and associated settings or locations

the X

No Comments

No Known Activations