INDEX

Explanations

words related to personal attributes or qualities

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "character" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

liga

-0.79

yg

-0.65

hner

-0.64

hem

-0.63

algia

-0.63

 Nunes

-0.63

IED

-0.62

Kop

-0.61

gard

-0.61

Ø§ÙĦ

-0.61

POSITIVE LOGITS

istics

1.28

acters

1.22

istically

1.20

izations

0.98

izes

0.95

 charact

0.95

isations

0.84

 assassination

0.79

anooga

0.79

isation

0.77

Activations Density 0.019%

words related to personal attributes or qualities

references to the concept of "character" in various contexts

No Comments

No Known Activations