INDEX

Explanations

various forms of grouping or categorization in text

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

mathematical and technical terms

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

ftagPool

-0.74

 مشين

-0.63

+#+

-0.56

évaluateur

-0.56

AutoField

-0.55

Tembelea

-0.55

 дописавши

-0.55

IsContent

-0.54

 виправивши

-0.54

клопе

-0.53

POSITIVE LOGITS

 duquel

0.38

ViewImports

0.38

 OMITBAD

0.36

 censiti

0.34

 seguida

0.33

 muligt

0.33

 parlant

0.32

 fieltro

0.32

 nombrado

0.31

 wobec

0.31

Activations Density 0.286%

various forms of grouping or categorization in text

mathematical and technical terms

No Comments

No Known Activations