INDEX

Explanations

descriptive references to specific geographical locations

oai_token-act-pair · gpt-3.5-turbo

the term "strip" in various contexts

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

 Waterloo

-0.76

BuyableInstoreAndOnline

-0.70

CI

-0.70

answered

-0.69

Osw

-0.69

 Ancest

-0.69

 Skills

-0.69

 Invention

-0.68

 advisors

-0.67

ypes

-0.67

POSITIVE LOGITS

 strip

4.01

 strips

2.95

 Strip

2.66

strip

2.51

 stripping

2.05

 stripped

1.59

 stripe

1.43

 Stri

1.36

stri

1.26

 striped

1.20

Activations Density 0.026%

descriptive references to specific geographical locations

the term "strip" in various contexts

No Comments

No Known Activations