INDEX

Explanations

phrases related to time durations, specifically mentioning weeks

oai_token-act-pair · gpt-3.5-turbo

mentions of time periods specifically related to "weeks"

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

ius

-0.79

iana

-0.69

iva

-0.68

hess

-0.65

oka

-0.61

upon

-0.61

enter

-0.61

grass

-0.60

 conserve

-0.60

ution

-0.59

POSITIVE LOGITS

 weeks

3.46

 months

2.74

 days

2.39

 Months

2.16

months

2.16

 Weeks

1.98

 years

1.95

 hours

1.94

 week

1.88

week

1.74

Activations Density 0.017%

phrases related to time durations, specifically mentioning weeks

mentions of time periods specifically related to "weeks"

No Comments

No Known Activations