INDEX

Explanations

phrases related to legal or official documents

oai_token-act-pair · gpt-3.5-turbo

references to legal documents or filings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Fortress

-0.75

 Sabres

-0.69

 Barbar

-0.69

 Belg

-0.68

 Kazakhstan

-0.68

 Monaco

-0.67

 Townsend

-0.67

ctica

-0.67

 CCTV

-0.63

 Clifford

-0.62

POSITIVE LOGITS

 writ

1.27

 Writ

0.93

writ

0.83

terness

0.82

eln

0.78

ritten

0.78

Writ

0.78

catentry

0.76

 petitions

0.74

atche

0.73

Activations Density 0.019%

phrases related to legal or official documents

references to legal documents or filings

No Comments

No Known Activations