INDEX

Explanations

government and regulatory terms, specifically related to Federal Communications and Election Commissions

oai_token-act-pair · gpt-3.5-turbo

references to federal regulatory bodies and their activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

agar

-0.67

bye

-0.66

ahime

-0.64

bom

-0.63

uckland

-0.63

abouts

-0.60

wcs

-0.59

vu

-0.59

akuya

-0.58

hillary

-0.58

POSITIVE LOGITS

 Commission

0.83

 Fighters

0.72

 Railroad

0.71

 Society

0.70

 Admission

0.66

 Simulator

0.66

etary

0.66

 Examination

0.66

 Branch

0.65

 Association

0.65

Activations Density 0.122%

government and regulatory terms, specifically related to Federal Communications and Election Commissions

references to federal regulatory bodies and their activities

No Comments

No Known Activations