INDEX

Explanations

official terms or keywords in a document

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.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

Ïī

-0.82

livion

-0.73

adr

-0.70

under

-0.70

ifice

-0.69

within

-0.69

whatever

-0.68

conservancy

-0.68

ggle

-0.68

assic

-0.68

POSITIVE LOGITS

 Thoughts

1.30

 Results

1.29

 Features

1.28

 Contents

1.26

 Locations

1.24

 Examples

1.23

 Notes

1.21

 Comments

1.19

 Description

1.18

 Details

1.16

Activations Density 0.169%

official terms or keywords in a document

No Comments

No Known Activations

official terms or keywords in a document

No Comments

No Known Activations