INDEX

Explanations

mentions of weekends

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "weekend."

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

eus

-0.83

metics

-0.83

ifice

-0.78

rator

-0.78

ration

-0.78

mitter

-0.78

abet

-0.77

rate

-0.75

rates

-0.75

cro

-0.73

POSITIVE LOGITS

 mornings

1.24

 afternoon

1.11

 morning

1.04

 nights

1.03

 weekend

1.02

 evenings

0.98

 evening

0.98

 brunch

0.93

 night

0.93

 weekends

0.92

Activations Density 0.015%

mentions of weekends

references to the concept of "weekend."

No Comments

No Known Activations