INDEX

Explanations

terms related to disabling or managing hardware functionality

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

 computadora

-0.37

EndInit

-0.34

 computer

-0.31

 élu

-0.30

une

-0.30

domésticos

-0.30

Israël

-0.30

 AssemblyProduct

-0.30

CreateModel

-0.30

 computadoras

-0.29

POSITIVE LOGITS

GEBURTSDATUM

0.60

 <<<<<<<<<<<<<<

0.59

发表于

0.59

contentLoaded

0.56

 Numerade

0.53

 الرياضيه

0.53

aarrggbb

0.52

sedown

0.50

Aholisi

0.50

 Italijani

0.48

Activations Density 0.058%

terms related to disabling or managing hardware functionality

No Comments

No Known Activations