INDEX

Explanations

phrases related to obtaining, possessing, or engaging with objects

oai_token-act-pair · gpt-3.5-turbo

instances of "some" related to various contexts or activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

SPONSORED

-0.95

âĢ¢âĢ¢

-0.73

urden

-0.72

istan

-0.67

ip

-0.66

gon

-0.65

ean

-0.65

uckle

-0.64

undle

-0.64

vest

-0.63

POSITIVE LOGITS

place

1.25

body

0.97

how

0.96

ones

0.94

 serious

0.83

 interesting

0.82

 semblance

0.81

 nifty

0.80

 notable

0.80

 additional

0.75

Activations Density 0.108%

phrases related to obtaining, possessing, or engaging with objects

instances of "some" related to various contexts or activities

No Comments

No Known Activations