INDEX

Explanations

the name "Bram" and related texts

oai_token-act-pair · gpt-3.5-turbo

names of individuals and brands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ebus

-0.74

iners

-0.68

"]=>

-0.68

ãħĭ

-0.67

olk

-0.66

icago

-0.65

ports

-0.65

ð

-0.65

ynasty

-0.64

ocene

-0.64

POSITIVE LOGITS

 Bram

1.14

ĸļ

0.99

 Vaughan

0.85

bolt

0.78

 Volt

0.73

 Blaz

0.72

bite

0.72

 Melvin

0.71

staking

0.69

bent

0.69

Activations Density 0.008%

the name "Bram" and related texts

names of individuals and brands

No Comments

No Known Activations