INDEX

Explanations

specific words related to catalogs or listings

oai_token-act-pair · gpt-3.5-turbo

references to various types of catalogs

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

Downloadha

-0.83

thritis

-0.68

awar

-0.68

maxwell

-0.67

 Johann

-0.67

wi

-0.66

atever

-0.64

outher

-0.63

adows

-0.62

twitch

-0.62

POSITIVE LOGITS

uing

1.20

ued

1.15

ues

1.07

 catalogue

1.06

 catalog

1.04

eers

0.98

ysis

0.89

urally

0.83

ãĤ¼ãĤ¦ãĤ¹

0.82

alogy

0.81

Activations Density 0.021%

specific words related to catalogs or listings

references to various types of catalogs

No Known Activations