INDEX

Explanations

professionals and experts in various fields

oai_token-act-pair · gpt-3.5-turbo

words that refer to groups of people or professions

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

 Territories

-0.66

 Kitchen

-0.64

CHAR

-0.60

 Bund

-0.59

 conviction

-0.57

 Connection

-0.57

 Farm

-0.56

 Warden

-0.56

 Thing

-0.54

ãĥĥ

-0.54

POSITIVE LOGITS

hip

1.20

hips

1.10

paces

1.00

mith

0.99

ervatives

0.97

uggest

0.95

ucker

0.87

 extraord

0.86

pace

0.85

Activations Density 0.112%

professionals and experts in various fields

words that refer to groups of people or professions

No Comments

No Known Activations