INDEX

Explanations

abbreviations related to different programs and institutions

oai_token-act-pair · gpt-3.5-turbo

references to specific programs or initiatives denoted by acronyms

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

advertising

-0.73

bons

-0.68

lio

-0.68

zai

-0.67

anwhile

-0.65

 WATCHED

-0.63

berman

-0.62

cules

-0.62

 unanim

-0.61

lda

-0.61

POSITIVE LOGITS

iamond

1.28

ependent

1.17

ynamic

1.15

iesel

1.14

ynam

1.14

ictionary

1.13

ynasty

1.12

ouble

1.10

iscovery

1.04

etermination

1.04

Activations Density 0.082%

abbreviations related to different programs and institutions

references to specific programs or initiatives denoted by acronyms

No Comments

No Known Activations