INDEX

Explanations

references to British institutions and notable entities

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

agency's U.S. clients

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

сылкі

-0.33

 analyzed

-0.32

 viability

-0.31

jelly

-0.31

 lavor

-0.31

 Laub

-0.29

 fieltro

-0.29

 Warte

-0.29

 depth

-0.29

원

-0.28

POSITIVE LOGITS

出版年

0.59

afficheront

0.56

WebElementEntity

0.53

ⓧ

0.51

tagext

0.51

postIndex

0.50

principalTable

0.49

Hozzáférés

0.48

 ComVisible

0.47

adaptiveStyles

0.46

Activations Density 0.094%

references to British institutions and notable entities

agency's U.S. clients

No Comments

No Known Activations