INDEX

Explanations

programming-related keywords and functions

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

addCriterion

-0.70

MockBean

-0.65

jsxFileName

-0.62

DockStyle

-0.62

PropertyChanging

-0.61

parsedMessage

-0.60

 textStatus

-0.60

Derbyniad

-0.59

 queſta

-0.59

@[

-0.58

POSITIVE LOGITS

 judiciaire

0.46

 Keuangan

0.44

 Lingkungan

0.38

 Usaha

0.36

legd

0.35

 brug

0.35

 juge

0.35

 sense

0.35

 nocturno

0.34

 privilégi

0.34

Activations Density 0.003%

programming-related keywords and functions

No Comments

No Known Activations