INDEX

Explanations

technical terms and references related to programming and academic citations

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 Monfieur

-0.76

 Verſ

-0.74

ViewFeatures

-0.72

webElementXpaths

-0.71

 ſont

-0.71

 myſelf

-0.70

 ſche

-0.69

 Meksiko

-0.67

 itſelf

-0.67

 ainfi

-0.66

POSITIVE LOGITS

 Portugal

0.52

Portugal

0.46

Bissau

0.41

 Portuguese

0.39

Portuguese

0.36

 portug

0.32

por

0.31

우

0.31

 portugal

0.31

bis

0.30

Activations Density 0.077%

technical terms and references related to programming and academic citations

No Comments

No Known Activations