INDEX

Explanations

instances of phrases with "the" followed by a specific noun

oai_token-act-pair · gpt-3.5-turbo

references to a specific location or position

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

fulness

-0.74

eers

-0.74

 firsthand

-0.72

ARA

-0.71

fully

-0.71

BIL

-0.70

lance

-0.70

aris

-0.68

ersen

-0.68

ãĤ´ãĥ³

-0.66

POSITIVE LOGITS

 proverbial

0.83

 smallest

0.83

 continent

0.81

 country

0.80

 circle

0.76

 spectrum

0.76

 pack

0.73

 globe

0.72

 envelope

0.71

 compass

0.70

Activations Density 0.181%

instances of phrases with "the" followed by a specific noun

references to a specific location or position

No Comments

No Known Activations