INDEX

Explanations

terms related to the internet service Google

oai_token-act-pair · gpt-3.5-turbo

terms related to government and governance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

catentry

-0.75

channelAvailability

-0.73

Pwr

-0.72

 COUR

-0.71

ĳå£«

-0.69

 guiActiveUn

-0.64

Pear

-0.62

-+-+

-0.60

 Pend

-0.60

ãĥł

-0.60

POSITIVE LOGITS

lers

0.74

uld

0.71

warm

0.71

arent

0.69

zyme

0.68

onz

0.67

anic

0.67

vernment

0.67

emaker

0.66

bles

0.65

Activations Density 0.054%

terms related to the internet service Google

terms related to government and governance

No Comments

No Known Activations