INDEX

Explanations

references to technical terms and data sources in scientific contexts

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

specific acronyms and company names

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

HomeAsUp

-0.39

₦

-0.35

明明

-0.33

 Photocase

-0.33

op

-0.32

évaluateur

-0.32

hopped

-0.32

主

-0.32

biet

-0.32

Atsauces

-0.31

POSITIVE LOGITS

 للمعارف

0.71

 الحره

0.59

MLLoader

0.57

 مرئيه

0.53

EndInit

0.49

 wikipagina

0.49

DockStyle

0.48

 científicas

0.48

bootstrapcdn

0.47

 imageNamed

0.47

Activations Density 0.078%

references to technical terms and data sources in scientific contexts

specific acronyms and company names

No Comments

No Known Activations