INDEX

Explanations

references to a particular organization or entity denoted as "GA" along with different numerical values

oai_token-act-pair · gpt-3.5-turbo

references to GA (General Assembly or another relevant organization)

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enegger

-1.01

lihood

-0.77

ology

-0.76

ologists

-0.72

 Chomsky

-0.71

landish

-0.70

 Corbyn

-0.69

entially

-0.68

 Kardashian

-0.68

Else

-0.68

POSITIVE LOGITS

GA

1.13

vernment

1.04

UGE

1.00

GA

0.95

ZA

0.92

FF

0.81

IL

0.80

VE

0.77

VA

0.76

BA

0.76

Activations Density 0.004%

references to a particular organization or entity denoted as "GA" along with different numerical values

references to GA (General Assembly or another relevant organization)

No Comments

No Known Activations