INDEX

Explanations

statistical significance indicators and p-values in research data analysis

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

p-value significance testing

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

 StatelessWidget

-0.59

BeginContext

-0.56

AndEndTag

-0.53

 محفوظة

-0.50

TagMode

-0.49

anmar

-0.47

 CreateTagHelper

-0.46

ThroughAttribute

-0.43

MessageOf

-0.42

EndInit

-0.42

POSITIVE LOGITS

 probabilidad

0.44

 samfun

0.43

unnitel

0.42

 fédéral

0.41

 impreso

0.41

 dégâts

0.39

 probability

0.39

uanto

0.39

 résultat

0.38

точници

0.38

Activations Density 0.332%

statistical significance indicators and p-values in research data analysis

p-value significance testing

No Comments

No Known Activations