INDEX

Explanations

references to blogs

oai_token-act-pair · gpt-3.5-turbo

references to "Blog."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Lauder

-0.74

TRUMP

-0.71

asonic

-0.68

IELD

-0.68

 Scotia

-0.67

arching

-0.65

BuyableInstoreAndOnline

-0.65

eded

-0.61

 foil

-0.58

 Winchester

-0.58

POSITIVE LOGITS

gers

1.27

ging

1.15

ger

1.11

 Blog

0.94

osphere

0.88

Blog

0.85

glers

0.85

ged

0.83

post

0.82

bub

0.82

Activations Density 0.009%

references to blogs

references to "Blog."

No Comments

No Known Activations