INDEX

Explanations

phrases related to books, research, and publications

oai_token-act-pair · gpt-3.5-turbo

phrases that denote authorship or refer to works and their titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 inhibitor

-0.84

 tether

-0.83

 queen

-0.83

 pumps

-0.81

 exhib

-0.79

 dipping

-0.78

 bolt

-0.76

 versa

-0.76

 tower

-0.75

 totem

-0.75

POSITIVE LOGITS

 Excellence

1.51

 Everyday

1.48

 Contemporary

1.41

 Practice

1.40

 Difference

1.40

 Aging

1.39

 Names

1.38

 Relations

1.38

 Organizations

1.38

 Context

1.38

Activations Density 0.236%

phrases related to books, research, and publications

phrases that denote authorship or refer to works and their titles

No Comments

No Known Activations