INDEX

Explanations

numerical values in a certain format, potentially related to dates, amounts, or measurements

oai_token-act-pair · gpt-3.5-turbo

numbers and numerical data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Vaugh

-0.65

 disadvant

-0.59

 referen

-0.59

ãĥ¼ãĥĨãĤ£

-0.56

 thous

-0.55

 predec

-0.52

 Moroc

-0.51

 destro

-0.51

 challeng

-0.49

 fundament

-0.49

POSITIVE LOGITS

taboola

0.52

rg

0.47

 COURT

0.47

 âĢº

0.47

HOU

0.45

mel

0.45

serv

0.45

 DARK

0.45

 DRAGON

0.45

 GOODMAN

0.44

Activations Density 0.981%

numerical values in a certain format, potentially related to dates, amounts, or measurements

numbers and numerical data

No Comments

No Known Activations