INDEX

Explanations

abbreviations and acronyms containing the letters "LI" or "I"

oai_token-act-pair · gpt-3.5-turbo

abbreviations or acronyms

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

 spare

-0.69

nets

-0.66

icate

-0.64

 puberty

-0.64

 altern

-0.62

icative

-0.60

ictive

-0.58

body

-0.57

 minimized

-0.57

ilaterally

-0.57

POSITIVE LOGITS

RECT

1.31

KE

1.21

OUS

1.16

YA

1.13

EG

1.11

LEY

1.11

AMI

1.10

RA

1.09

EEE

1.06

BA

1.05

Activations Density 0.066%

abbreviations and acronyms containing the letters "LI" or "I"

abbreviations or acronyms

No Comments

No Known Activations