INDEX

Explanations

statements related to endorsements and opinions

oai_token-act-pair · gpt-3.5-turbo

phrases related to social and political engagement

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 thereof

-0.95

 thereto

-0.77

Firstly

-0.74

Interstitial

-0.73

 afterwards

-0.68

They

-0.68

viation

-0.68

 unlucky

-0.67

Its

-0.64

arers

-0.64

POSITIVE LOGITS

 Coulter

1.03

 DeVos

1.02

 Scalia

1.00

Doe

0.97

 Congressman

0.96

 Huckabee

0.95

 Yiannopoulos

0.94

 Gorsuch

0.94

 Mueller

0.93

 ALEC

0.92

Activations Density 0.907%

statements related to endorsements and opinions

phrases related to social and political engagement

No Comments

No Known Activations