INDEX

Explanations

acronyms related to U.S. government agencies

oai_token-act-pair · gpt-3.5-turbo

mentions of government agencies or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lihood

-0.91

 Australians

-0.79

 Belg

-0.78

acters

-0.73

 WhatsApp

-0.73

 Corbyn

-0.72

Austral

-0.71

 Samurai

-0.69

 Britons

-0.69

 Ãĸ

-0.69

POSITIVE LOGITS

 Administrator

0.97

 Secretary

0.83

 administrator

0.83

HQ

0.81

 secretary

0.81

ENCY

0.81

 departments

0.79

 commissioner

0.79

 whistleblower

0.79

 payroll

0.77

Activations Density 0.059%

acronyms related to U.S. government agencies

mentions of government agencies or organizations

No Comments

No Known Activations