INDEX

Explanations

terms relating to legal or official documents

oai_token-act-pair · gpt-3.5-turbo

legal terminology and references to legal cases or documents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

itably

-0.82

ancial

-0.77

irlf

-0.77

escription

-0.73

;;;;;;;;;;;;

-0.73

ividual

-0.70

ividually

-0.70

retty

-0.70

mercial

-0.69

oided

-0.68

POSITIVE LOGITS

CLASSIFIED

1.08

NetMessage

0.96

WAYS

0.87

 behavi

0.85

natureconservancy

0.83

 largeDownload

0.82

 toile

0.80

GoldMagikarp

0.78

 welf

0.76

Balt

0.72

Activations Density 4.596%

terms relating to legal or official documents

legal terminology and references to legal cases or documents

No Comments

No Known Activations