INDEX

Explanations

names of people, places, and organizations

oai_token-act-pair · gpt-3.5-turbo

specific words and phrases related to numerical and statistical data or classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 individually

-0.48

 SHALL

-0.45

 Cosponsors

-0.42

NON

-0.42

 includ

-0.42

 downstream

-0.41

 Tuls

-0.41

 sidx

-0.41

 guiName

-0.40

 GOODMAN

-0.40

POSITIVE LOGITS

cour

0.55

cyclop

0.52

elligence

0.47

 fame

0.46

croft

0.46

econom

0.45

castle

0.45

dar

0.45

pres

0.44

anus

0.44

Activations Density 1.844%

names of people, places, and organizations

specific words and phrases related to numerical and statistical data or classifications

No Comments

No Known Activations