INDEX

Explanations

phrases related to providing or being provided by someone or something

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate the source or attribution of information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ivia

-0.81

bra

-0.76

enth

-0.73

abouts

-0.73

hemer

-0.72

weed

-0.70

gregation

-0.69

ÃŃ

-0.69

ices

-0.68

ifer

-0.67

POSITIVE LOGITS

 virtue

0.96

laws

0.93

 statute

0.79

products

0.76

 afar

0.75

 fiat

0.71

 multiplying

0.69

 Almighty

0.69

 omission

0.67

 means

0.66

Activations Density 0.140%

phrases related to providing or being provided by someone or something

phrases that indicate the source or attribution of information

No Comments

No Known Activations