INDEX

Explanations

items or categories in a list

oai_token-act-pair · gpt-3.5-turbo

references to products, tracks, or projects related to entertainment or media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 goddamn

-0.66

yet

-0.65

paces

-0.64

yond

-0.64

 godd

-0.62

',"

-0.62

ghan

-0.60

––

-0.60

 damned

-0.59

 legion

-0.58

POSITIVE LOGITS

 Description

1.53

 Details

1.50

 Overview

1.42

 Contents

1.36

 Summary

1.32

 Requirements

1.32

 Highlights

1.32

 Notes

1.32

 Specifications

1.29

 Features

1.28

Activations Density 0.302%

items or categories in a list

references to products, tracks, or projects related to entertainment or media

No Comments

No Known Activations