INDEX

Explanations

mentions of political positions or governmental roles

oai_token-act-pair · gpt-3.5-turbo

mentions of governors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

opter

-0.81

itton

-0.78

ramid

-0.71

origin

-0.71

ener

-0.69

andem

-0.69

pos

-0.68

Else

-0.68

bra

-0.67

dest

-0.67

POSITIVE LOGITS

 governor

1.03

 Governor

1.00

 Cuomo

0.84

 governors

0.82

 Governors

0.76

govern

0.74

Gov

0.73

 Schwarzenegger

0.72

 antid

0.71

 Mansion

0.71

Activations Density 0.011%

mentions of political positions or governmental roles

mentions of governors

No Comments

No Known Activations