INDEX

Explanations

words related to specific terms, including technical jargon, names, and numerical quantities

oai_token-act-pair · gpt-3.5-turbo

terminology related to specific identities, rules, or classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

PROV

-0.59

osaurs

-0.58

 Copyright

-0.54

Pg

-0.53

natureconservancy

-0.51

Flickr

-0.50

halla

-0.50

ãĢ

-0.50

::::::::

-0.50

================================================================

-0.49

POSITIVE LOGITS

rique

0.68

eers

0.65

 pregn

0.59

alion

0.57

rul

0.57

 skelet

0.56

ificate

0.56

 Cooke

0.52

nered

0.51

 blat

0.50

Activations Density 1.530%

words related to specific terms, including technical jargon, names, and numerical quantities

terminology related to specific identities, rules, or classifications

No Comments

No Known Activations