INDEX

Explanations

characters from a specific encoding system

oai_token-act-pair · gpt-3.5-turbo

symbols or characters used in special contexts, likely related to non-Latin scripts or encoding issues

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

 contrace

-0.81

eatures

-0.78

orno

-0.77

 mathemat

-0.76

abase

-0.74

icably

-0.72

 coils

-0.71

wana

-0.71

glomer

-0.70

 implants

-0.70

POSITIVE LOGITS

Ø±

1.10

ÙĦ

1.09

ÙĴ

1.04

ÙĲ

1.03

à¤¾

1.02

Øª

1.01

ÙĬ

1.01

Ùİ

0.97

à¥

0.95

Ø¨

0.93

Activations Density 0.018%

characters from a specific encoding system

symbols or characters used in special contexts, likely related to non-Latin scripts or encoding issues

No Comments

No Known Activations