INDEX

Explanations

symbols and special characters often related to formatting or encoding

oai_token-act-pair · gpt-3.5-turbo

references to statistical data or metrics related to demographics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

boro

-0.73

 favorably

-0.69

 favor

-0.69

 endeavors

-0.69

pei

-0.68

ĸļ

-0.66

agos

-0.66

 honors

-0.66

yrights

-0.66

¥µ

-0.65

POSITIVE LOGITS

However

0.90

Labour

0.90

Scotland

0.88

England

0.88

Meanwhile

0.86

Britain

0.83

Refer

0.80

Australian

0.79

Topics

0.79

Shape

0.79

Activations Density 0.445%

symbols and special characters often related to formatting or encoding

references to statistical data or metrics related to demographics

No Comments

No Known Activations