INDEX

Explanations

phrases related to specific names, particularly "Sandra" and "Chandra"

oai_token-act-pair · gpt-3.5-turbo

references to the name "Sandra" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

cffff

-0.99

¥ŀ

-0.91

eatures

-0.79

olesc

-0.74

onet

-0.74

 tradem

-0.72

icipated

-0.71

ategory

-0.70

 affili

-0.70

natureconservancy

-0.69

POSITIVE LOGITS

andra

1.14

 Sandra

1.04

 Bland

0.93

ãĤ©

0.80

BuyableInstoreAndOnline

0.80

 Ellen

0.77

 Grace

0.76

nce

0.75

 Maria

0.74

 Flores

0.73

Activations Density 0.010%

phrases related to specific names, particularly "Sandra" and "Chandra"

references to the name "Sandra" and its variations

No Comments

No Known Activations