INDEX

Explanations

quantitative information related to various metrics such as numbers, percentages, and figures

oai_token-act-pair · gpt-3.5-turbo

data related to statistics and numerical trends in various contexts

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

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

 Canaver

-0.68

 Doing

-0.62

fman

-0.59

 Penal

-0.58

entious

-0.58

cause

-0.56

rament

-0.55

rir

-0.55

 ILCS

-0.55

 Dund

-0.54

POSITIVE LOGITS

 varies

1.55

 exceeds

1.38

 outweigh

1.26

 depends

1.25

 differs

1.24

 dwind

1.22

 vary

1.21

 decreased

1.20

 differed

1.15

 increased

1.15

Activations Density 0.358%

quantitative information related to various metrics such as numbers, percentages, and figures

data related to statistics and numerical trends in various contexts

No Comments

No Known Activations