INDEX

Explanations

technical terms and indicators related to scientific measurements and processes

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

between or comparison

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

AnchorStyles

-0.66

PerformLayout

-0.54

principalColumn

-0.54

StructEnd

-0.52

invokeLater

-0.52

 TestBed

-0.51

يكب

-0.50

 AssemblyCulture

-0.50

Portail

-0.49

 OnInit

-0.49

POSITIVE LOGITS

 between

0.41

 antara

0.36

GEBURTSDATUM

0.36

 обра

0.34

 among

0.34

 zwischen

0.33

ungsver

0.33

 rapporti

0.32

 BETWEEN

0.32

 Beziehungen

0.31

Activations Density 0.195%

technical terms and indicators related to scientific measurements and processes

between or comparison

No Comments

No Known Activations