INDEX

Explanations

terms related to enclosed spaces or barriers

oai_token-act-pair · gpt-3.5-turbo

terms related to enclosures and contained spaces

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

nant

-0.81

hist

-0.72

zinski

-0.69

bers

-0.69

common

-0.67

Crunch

-0.67

sonian

-0.67

enthal

-0.67

 Mysteries

-0.67

ann

-0.66

POSITIVE LOGITS

 encl

1.50

osures

1.39

 enclosure

1.23

 enclosed

1.02

izabeth

0.83

 cages

0.81

 spaces

0.80

 rooms

0.79

ridges

0.78

 surrounds

0.74

Activations Density 0.026%

terms related to enclosed spaces or barriers

terms related to enclosures and contained spaces

No Comments

No Known Activations