INDEX

Explanations

words related to intentions or motivations

oai_token-act-pair · gpt-3.5-turbo

infinitive phrases indicating intention or desire

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

soDeliveryDate

-0.89

listed

-0.72

calling

-0.70

warning

-0.67

 Printed

-0.67

 listings

-0.66

details

-0.65

case

-0.64

 Case

-0.63

measures

-0.63

POSITIVE LOGITS

 participate

1.33

 behave

1.32

 engage

1.31

 obey

1.29

 perform

1.28

 consume

1.27

 partake

1.25

eat

1.23

 achieve

1.20

 pursue

1.19

Activations Density 0.383%

words related to intentions or motivations

infinitive phrases indicating intention or desire

No Comments

No Known Activations