INDEX

Explanations

references to a specific person named "Arafat"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Arafat" and related figures

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

eners

-0.83

ership

-0.82

bilt

-0.76

Legendary

-0.68

Vector

-0.67

tainment

-0.67

liness

-0.67

gamer

-0.66

sylvania

-0.66

itution

-0.64

POSITIVE LOGITS

Ara

1.03

ikawa

0.90

ishi

0.87

uses

0.85

yip

0.79

rison

0.78

azel

0.77

byss

0.77

iries

0.76

 Palestin

0.75

Activations Density 0.017%

references to a specific person named "Arafat"

mentions of the name "Arafat" and related figures

No Comments

No Known Activations