INDEX

Explanations

phrases related to different entities or organizations

oai_token-act-pair · gpt-3.5-turbo

mentions of specific organizations, brands, or initiatives

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

wcs

-0.65

aying

-0.60

piring

-0.59

 metic

-0.58

 mathemat

-0.58

.ãĢį

-0.58

arial

-0.58

 surpr

-0.57

 conserv

-0.57

imar

-0.56

POSITIVE LOGITS

 Foundation

0.74

 Reborn

0.72

 Labs

0.72

 Development

0.72

eers

0.69

 itself

0.68

 Investigations

0.67

 Forum

0.67

folios

0.66

 Presents

0.65

Activations Density 0.409%

phrases related to different entities or organizations

mentions of specific organizations, brands, or initiatives

No Comments

No Known Activations