INDEX

Explanations

keywords related to legal and regulatory texts

oai_token-act-pair · gpt-3.5-turbo

legal references and sections from regulatory texts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 savior

-0.69

glers

-0.68

eton

-0.67

arin

-0.63

inka

-0.63

tones

-0.60

ammed

-0.59

 yarn

-0.59

GGGGGGGG

-0.58

 garg

-0.58

POSITIVE LOGITS

verning

0.96

Â§

0.83

Accessory

0.79

£ı

0.78

initions

0.77

Section

0.76

 Definitions

0.75

Article

0.74

 subsection

0.74

 Appendix

0.74

Activations Density 0.102%

keywords related to legal and regulatory texts

legal references and sections from regulatory texts

No Comments

No Known Activations