INDEX

Explanations

words related to causality and explanation

oai_token-act-pair · gpt-3.5-turbo

phrases indicating causation or reasoning

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

NetMessage

-0.45

natureconservancy

-0.37

ccording

-0.32

Pwr

-0.32

Eye

-0.32

 medic

-0.31

bard

-0.31

talk

-0.31

aed

-0.30

enei

-0.30

POSITIVE LOGITS

ividual

0.46

idth

0.42

éĹĺ

0.37

uart

0.36

hower

0.36

ocument

0.33

ierre

0.33

okane

0.33

 lasted

0.33

elvet

0.33

Activations Density 0.687%

words related to causality and explanation

phrases indicating causation or reasoning

No Comments

No Known Activations