INDEX

Explanations

function names and variable references in a programming context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.12.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

니다

-0.65

MathML

-0.63

Cubit

-0.59

 OMITBAD

-0.59

inigungs

-0.59

клопе

-0.58

 olympiques

-0.57

typeorm

-0.57

ViewImports

-0.57

OPP

-0.56

POSITIVE LOGITS

 estekak

0.69

 Inland

0.64

 Coch

0.64

 isos

0.61

تقاوى

0.59

 مشارکت‌کنندگان

0.59

uyện

0.58

 Mula

0.58

 فريبيس

0.57

 stell

0.57

Activations Density 0.013%

function names and variable references in a programming context

No Comments

No Known Activations