INDEX

Explanations

keywords related to names and specific terms, potentially related to legal cases or news pieces

oai_token-act-pair · gpt-3.5-turbo

mentions of specific family names and their associations in a legal context

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

Yel

-0.71

Paw

-0.71

 calendars

-0.70

Xia

-0.69

 fing

-0.65

Ys

-0.65

 paws

-0.64

 world

-0.64

 Tian

-0.63

Yam

-0.62

POSITIVE LOGITS

arson

4.66

osi

1.42

ramer

1.00

azar

0.95

lein

0.92

esson

0.91

sson

0.90

ersen

0.90

ovi

0.88

anson

0.87

Activations Density 0.028%

keywords related to names and specific terms, potentially related to legal cases or news pieces

mentions of specific family names and their associations in a legal context

No Comments

No Known Activations