INDEX

Explanations

words related to a particular location or person named "Raf"

oai_token-act-pair · gpt-3.5-turbo

the name "Raf" and its variations in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

chi

-0.86

ioch

-0.85

 Wyoming

-0.75

minecraft

-0.72

Boston

-0.69

church

-0.68

 Cherokee

-0.67

CHA

-0.66

Folder

-0.65

Kin

-0.65

POSITIVE LOGITS

Raf

4.14

 Rafael

2.46

 Maur

1.22

Riy

1.10

 Mous

1.08

 Houth

1.08

Nad

1.08

 Lowry

1.07

raf

1.07

Ara

1.06

Activations Density 0.041%

words related to a particular location or person named "Raf"

the name "Raf" and its variations in various contexts

No Comments

No Known Activations