INDEX

Explanations

words written with apostrophes before a number or letter

oai_token-act-pair · gpt-3.5-turbo

instances of quotation marks or phrases enclosed within quotation marks

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Leilan

-0.76

avorite

-0.76

âĹ¼

-0.73

 Izan

-0.70

 Archdemon

-0.67

 Mechdragon

-0.65

igslist

-0.64

 Journals

-0.64

nown

-0.63

 Hendricks

-0.63

POSITIVE LOGITS

til

1.15

cause

1.10

neath

1.07

',

1.04

-'

1.02

tis

0.97

Interstitial

0.97

Mech

0.93

'[

0.91

thumbnails

0.89

Activations Density 0.033%

words written with apostrophes before a number or letter

instances of quotation marks or phrases enclosed within quotation marks

No Comments

No Known Activations