INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

punctuation and specific character patterns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

anwhile

-0.81

eteria

-0.74

Downloadha

-0.71

abus

-0.70

agy

-0.69

apest

-0.69

risome

-0.68

gerald

-0.68

ossibility

-0.67

illions

-0.67

POSITIVE LOGITS

ç¥ŀ

0.72

 Doyle

0.70

 Vengeance

0.67

cel

0.65

ãĥı

0.64

 Byzantine

0.64

quez

0.63

Hes

0.63

 Sanct

0.62

 Diesel

0.61

Activations Density 0.383%

proper nouns

punctuation and specific character patterns

No Comments

No Known Activations