INDEX

Explanations

references to worshippers

oai_token-act-pair · gpt-3.5-turbo

words related to worshipers or worship

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 stage

-0.76

 Sense

-0.68

Zah

-0.64

 Stage

-0.63

 inst

-0.63

 stack

-0.61

HL

-0.60

Yar

-0.60

ethe

-0.60

stage

-0.59

POSITIVE LOGITS

ippers

4.38

ipper

2.87

ipping

2.00

ipped

1.53

ipp

1.52

appers

1.46

ips

1.41

ippy

1.40

IPP

1.20

IPS

1.19

Activations Density 0.005%

references to worshippers

words related to worshipers or worship

No Comments

No Known Activations