INDEX

Explanations

news or current events related phrases

oai_token-act-pair · gpt-3.5-turbo

phrases indicating occurrences or actions related to events or announcements

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

è¦ļéĨĴ

-0.72

ivo

-0.57

peer

-0.56

roots

-0.55

oller

-0.54

gom

-0.54

Advertisements

-0.54

itcher

-0.54

TEXTURE

-0.54

ynasty

-0.52

POSITIVE LOGITS

 amid

1.30

 amidst

1.14

 weeks

1.00

 days

1.00

 after

0.97

 shortly

0.95

 months

0.93

 hours

0.91

 despite

0.90

Days

0.84

Activations Density 0.044%

news or current events related phrases

phrases indicating occurrences or actions related to events or announcements

No Known Activations