INDEX

Explanations

phrases related to directories and folders

oai_token-act-pair · gpt-3.5-turbo

terms related to file system organization and directory structures

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

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

acting

-0.75

acted

-0.75

pter

-0.74

iating

-0.69

 Overt

-0.69

akening

-0.68

uca

-0.68

Ò

-0.65

oria

-0.65

immune

-0.65

POSITIVE LOGITS

 folder

0.93

ystem

0.91

 folders

0.91

'/

0.91

(/

0.90

~/.

0.89

 Contents

0.89

"/

0.87

 Folder

0.86

 contents

0.86

Activations Density 0.056%

phrases related to directories and folders

terms related to file system organization and directory structures

No Comments

No Known Activations