INDEX

Explanations

phrases related to bringing an end or resolution to something

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

DragonMagazine

-0.80

thouse

-0.67

rients

-0.67

hooting

-0.65

 toile

-0.64

æ©Ł

-0.63

WER

-0.63

 constitu

-0.62

hawk

-0.62

*/(

-0.61

POSITIVE LOGITS

angering

1.19

owment

1.05

orses

1.00

 hostilities

1.00

 slavery

0.97

ear

0.90

 homelessness

0.88

eared

0.88

 dependence

0.81

 reliance

0.80

Activations Density 0.045%

phrases related to bringing an end or resolution to something

No Comments

No Known Activations