INDEX

Explanations

the word "Me"

oai_token-act-pair · gpt-3.5-turbo

specific names and unique identifiers related to individuals or entities

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

LEG

-0.82

 Newman

-0.82

 Reyes

-0.80

 Legends

-0.78

iel

-0.78

 Kissinger

-0.78

MO

-0.77

RL

-0.75

 Lori

-0.73

 Kramer

-0.72

POSITIVE LOGITS

AT

1.48

ats

1.48

at

1.36

rat

1.33

atted

1.26

mat

1.18

bat

1.18

atus

1.17

Mat

1.16

pat

1.16

Activations Density 0.312%

the word "Me"

specific names and unique identifiers related to individuals or entities

No Comments

No Known Activations