INDEX

Explanations

information related to articles, including authors and editing notes

oai_token-act-pair · gpt-3.5-turbo

greetings and references to content structure or metadata in documents

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

forth

-0.61

 helicop

-0.61

stood

-0.60

 horizont

-0.58

dn

-0.57

aturdays

-0.55

pheus

-0.54

atever

-0.54

ilaterally

-0.53

akespe

-0.53

POSITIVE LOGITS

 âĢº

0.68

 SHARES

0.66

 largeDownload

0.63

UTERS

0.62

 Entered

0.59

 Expand

0.57

actionDate

0.56

BILITIES

0.54

=>

0.53

 Coverage

0.53

Activations Density 0.497%

information related to articles, including authors and editing notes

greetings and references to content structure or metadata in documents

No Comments

No Known Activations