INDEX

Explanations

phrases or acronyms related to specific programs or operations

oai_token-act-pair · gpt-3.5-turbo

references to "SP" or similar acronyms and their context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

omorph

-0.66

adm

-0.66

ebted

-0.65

 palm

-0.64

court

-0.64

isman

-0.62

hower

-0.62

Ã§a

-0.61

taboola

-0.61

omorphic

-0.61

POSITIVE LOGITS

EED

1.09

OIL

1.07

SP

1.05

ACE

1.01

ARK

1.00

ORTS

0.92

ADA

0.91

RIPT

0.90

IRT

0.89

ORT

0.89

Activations Density 0.006%

phrases or acronyms related to specific programs or operations

references to "SP" or similar acronyms and their context

No Comments

No Known Activations