INDEX

Explanations

numerical values such as dates, measurements, or rankings

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 nonetheless

-0.76

 likewise

-0.75

Such

-0.73

itto

-0.72

roo

-0.68

idious

-0.68

these

-0.68

ourke

-0.67

 lest

-0.66

utterstock

-0.65

POSITIVE LOGITS

GSL

0.71

NXT

0.68

 independ

0.66

 Emblem

0.66

 Protoss

0.64

 roadmap

0.60

 Ethereum

0.59

ORPG

0.59

 Astral

0.59

HDD

0.58

Activations Density 1.704%

numerical values such as dates, measurements, or rankings

No Comments

No Known Activations