INDEX

Explanations

specific locations or points where two or more things intersect

oai_token-act-pair · gpt-3.5-turbo

references to intersections

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

warm

-0.77

yss

-0.70

vation

-0.69

apons

-0.68

astic

-0.67

asp

-0.66

amina

-0.65

unker

-0.64

ULE

-0.63

kowski

-0.63

POSITIVE LOGITS

 intersection

1.44

 intersections

1.24

 intersect

1.00

 junction

0.88

 Frie

0.86

 Beir

0.76

 Sakuya

0.74

 crossings

0.71

 nexus

0.71

 mund

0.70

Activations Density 0.011%

specific locations or points where two or more things intersect

references to intersections

No Comments

No Known Activations