INDEX

Explanations

the word "fountain" at various activations

oai_token-act-pair · gpt-3.5-turbo

references to fountains

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

 ACTIONS

-0.65

VICE

-0.65

 WATCHED

-0.64

eon

-0.64

 Debor

-0.62

alien

-0.61

 Anarchy

-0.59

Ey

-0.59

nd

-0.59

 Cosponsors

-0.59

POSITIVE LOGITS

 fountain

1.28

 Fountain

1.27

 pens

0.87

issance

0.81

hesda

0.80

 empt

0.78

pen

0.77

sonian

0.76

tailed

0.73

shire

0.73

Activations Density 0.004%

the word "fountain" at various activations

references to fountains

No Comments

No Known Activations