INDEX

Explanations

proper nouns or names, particularly ones with the word "Rash" in them

oai_token-act-pair · gpt-3.5-turbo

names and references to individual athletes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

insula

-0.83

 Lauder

-0.74

 MacArthur

-0.67

éļ

-0.67

omore

-0.66

 Strait

-0.65

EXT

-0.64

 Springs

-0.63

 Hitman

-0.63

DCS

-0.62

POSITIVE LOGITS

interstitial

1.00

senal

0.98

uled

0.97

uling

0.82

alph

0.80

outine

0.76

dies

0.73

login

0.72

der

0.72

alos

0.67

Activations Density 0.031%

proper nouns or names, particularly ones with the word "Rash" in them

names and references to individual athletes

No Comments

No Known Activations