INDEX

Explanations

phrases indicating location or direction

oai_token-act-pair · gpt-3.5-turbo

phrases indicating connections or relationships, particularly those involving energy and the term "and."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

owered

-0.70

iple

-0.69

anship

-0.67

 Disability

-0.66

NI

-0.63

 Detection

-0.62

omy

-0.60

DragonMagazine

-0.60

 Cognitive

-0.59

ZE

-0.59

POSITIVE LOGITS

 elsewhere

1.34

 here

1.31

 abroad

1.21

here

1.09

 there

1.07

 HERE

1.06

 hereafter

1.04

 nowhere

1.04

 everywhere

0.99

 THERE

0.88

Activations Density 0.055%

phrases indicating location or direction

phrases indicating connections or relationships, particularly those involving energy and the term "and."

No Comments

No Known Activations