INDEX

Explanations

references to supplementary materials and figures in research documents

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

academic references and citations

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

 iNdEx

-0.51

AnchorTagHelper

-0.51

+#+

-0.50

ftagPool

-0.49

 мәкал

-0.48

riwal

-0.48

 Superhost

-0.47

的笑容

-0.47

ğından

-0.46

 ASSISTANCE

-0.46

POSITIVE LOGITS

PhysRev

0.53

Jeografia

0.46

 nakalista

0.41

 createContext

0.39

 acca

0.38

Démographie

0.37

PhysRevD

0.36

晦

0.36

-------------</

0.35

WebServlet

0.35

Activations Density 0.077%

references to supplementary materials and figures in research documents

academic references and citations

No Comments

No Known Activations