INDEX

Explanations

instances of the word "Phoenicians"

oai_token-act-pair · gpt-3.5-turbo

names or terms associated with specific locations or populations

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

izoph

-0.80

Ĥª

-0.73

Pwr

-0.71

HAEL

-0.67

 Ambro

-0.64

 Manila

-0.64

RED

-0.63

ENTS

-0.62

Jet

-0.62

dos

-0.62

POSITIVE LOGITS

vironment

1.48

zyme

1.06

esis

1.02

viron

0.91

vironments

0.88

uve

0.86

isance

0.85

pard

0.85

cedes

0.81

itent

0.81

Activations Density 0.019%

instances of the word "Phoenicians"

names or terms associated with specific locations or populations

No Comments

No Known Activations