INDEX

Explanations

phrases related to belief statements

oai_token-act-pair · gpt-3.5-turbo

phrases that include the expression "believe it or not."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

istries

-0.78

 Inventory

-0.70

pload

-0.68

 Catalog

-0.65

 Baldwin

-0.64

ordes

-0.61

anchester

-0.60

æ©Ł

-0.60

shire

-0.60

ories

-0.59

POSITIVE LOGITS

epad

1.01

hin

0.83

MpServer

0.78

 necessarily

0.71

ched

0.70

ches

0.70

hood

0.66

 recip

0.65

ionic

0.65

cooked

0.65

Activations Density 0.061%

phrases related to belief statements

phrases that include the expression "believe it or not."

No Comments

No Known Activations