INDEX

Explanations

elements related to official statements or decisions

oai_token-act-pair · gpt-3.5-turbo

expressions related to political statements or actions

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

Todd

-0.84

Madison

-0.72

 Lovecraft

-0.70

 Whitney

-0.67

Whit

-0.67

Ohio

-0.67

 Simmons

-0.67

 Shelby

-0.65

Chuck

-0.65

 Garrett

-0.64

POSITIVE LOGITS

NGO

0.87

 organis

0.87

 Azerbai

0.87

 authorised

0.83

 organise

0.82

 Azerbaijan

0.80

 NGOs

0.80

 UNESCO

0.80

Mub

0.80

 Ankara

0.79

Activations Density 3.969%

elements related to official statements or decisions

expressions related to political statements or actions

No Comments

No Known Activations