INDEX

Explanations

references to legal and judicial matters

oai_token-act-pair · gpt-3.5-turbo

references to court rulings and judicial opinions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 destro

-0.80

obyl

-0.73

 livest

-0.73

 sacrific

-0.73

 Kardashian

-0.72

 Kurd

-0.71

uilt

-0.70

 pent

-0.67

 Pesh

-0.67

 Sunrise

-0.67

POSITIVE LOGITS

 justices

1.04

 Scalia

0.92

rooms

0.85

Justice

0.83

room

0.80

sup

0.79

 Gorsuch

0.76

pedia

0.76

Layout

0.74

 Clarence

0.73

Activations Density 0.059%

references to legal and judicial matters

references to court rulings and judicial opinions

No Comments

No Known Activations