INDEX

Explanations

statements made by prominent individuals or figures

oai_token-act-pair · gpt-3.5-turbo

instances of speech or quotes by individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Newsletter

-1.01

Cf

-0.74

Bal

-0.67

jam

-0.66

osc

-0.65

âĹ¼

-0.63

ItemImage

-0.63

_-

-0.62

Mech

-0.62

wk

-0.60

POSITIVE LOGITS

 homosexuality

0.80

 passionately

0.72

 Scientology

0.72

 embattled

0.69

 plans

0.67

 homosexuals

0.66

 constitu

0.65

 euth

0.65

gay

0.64

 aloud

0.64

Activations Density 0.325%

statements made by prominent individuals or figures

instances of speech or quotes by individuals

No Comments

No Known Activations