INDEX

Explanations

phrases related to familiarity or common knowledge

oai_token-act-pair · gpt-3.5-turbo

phrases or expressions indicating familiarity or common experiences

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

 Accessed

-0.72

sterdam

-0.72

 smokes

-0.67

 largeDownload

-0.65

afety

-0.63

uli

-0.63

ahead

-0.62

backs

-0.61

croft

-0.60

MORE

-0.60

POSITIVE LOGITS

ggles

1.00

ilet

0.74

pper

0.74

 contemplate

0.73

 outsiders

0.73

wered

0.72

 behold

0.71

asty

0.70

ADS

0.70

 everyone

0.70

Activations Density 0.178%

phrases related to familiarity or common knowledge

phrases or expressions indicating familiarity or common experiences

No Comments

No Known Activations