INDEX

Explanations

names related to politics and potential legal actions

oai_token-act-pair · gpt-3.5-turbo

identifiers and metrics related to individuals and entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orem

-0.71

ropolis

-0.64

edIn

-0.60

eus

-0.59

planet

-0.58

iveness

-0.55

arians

-0.55

eed

-0.54

hed

-0.54

 Princ

-0.53

POSITIVE LOGITS

¼

0.63

IN

0.59

ãĥŃ

0.57

ine

0.57

INE

0.56

in

0.54

¾

0.52

Ã»

0.52

é¾įå

0.51

¨

0.51

Activations Density 0.379%

names related to politics and potential legal actions

identifiers and metrics related to individuals and entities

No Comments

No Known Activations