INDEX

Explanations

mentions of improvement or positive change

oai_token-act-pair · gpt-3.5-turbo

terms related to improvement and progress

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

leton

-0.69

cium

-0.68

 Straw

-0.68

pull

-0.66

lua

-0.63

thur

-0.63

cia

-0.62

NetMessage

-0.61

Tumblr

-0.61

í

-0.61

POSITIVE LOGITS

ments

0.89

ainer

0.78

iets

0.78

heastern

0.78

 grades

0.76

 improve

0.72

 improves

0.71

 enhancement

0.71

cyclopedia

0.70

atson

0.70

Activations Density 0.028%

mentions of improvement or positive change

terms related to improvement and progress

No Comments

No Known Activations