INDEX

Explanations

Twitter usernames or handles

oai_token-act-pair · gpt-3.5-turbo

abbreviations and acronyms commonly related to internet and coding contexts

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

 Galile

-0.79

ãĥ¼ãĥĨ

-0.57

 Tata

-0.57

Sci

-0.57

 incentive

-0.56

 embell

-0.56

 VIDE

-0.56

 Heist

-0.55

 Bucks

-0.55

 watt

-0.55

POSITIVE LOGITS

ullivan

1.56

nyder

1.55

anchez

1.48

erver

1.48

ystem

1.45

ource

1.41

pace

1.41

hift

1.41

leep

1.41

ilver

1.40

Activations Density 0.123%

Twitter usernames or handles

abbreviations and acronyms commonly related to internet and coding contexts

No Comments

No Known Activations