INDEX

Explanations

transitional phrases and key terms that signal important information or data points

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Personendaten

-0.40

waz

-0.32

 grossen

-0.31

 deseado

-0.30

envie

-0.30

色々

-0.30

ام

-0.30

 alliés

-0.30

loo

-0.30

oOo

-0.29

POSITIVE LOGITS

帖最后由

0.63

 surla

0.59

 AssemblyCompany

0.58

protoimpl

0.57

ंदीखरीदारी

0.57

ViewImports

0.56

DockStyle

0.54

VersionUID

0.53

 MainAxisSize

0.53

transQ

0.52

Activations Density 0.244%

transitional phrases and key terms that signal important information or data points

No Comments

No Known Activations