INDEX

Explanations

mentions of news or information

oai_token-act-pair · gpt-3.5-turbo

phrases that begin with "news of" followed by significant events or occurrences

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ertodd

-0.74

hesis

-0.68

 ILCS

-0.68

nets

-0.67

olesterol

-0.64

aceous

-0.64

allery

-0.63

othal

-0.63

ulla

-0.62

zers

-0.62

POSITIVE LOGITS

 impending

0.95

 condol

0.72

enance

0.71

 announcing

0.70

 wrongdoing

0.69

 sorts

0.68

 imminent

0.68

 yesterday

0.67

 scant

0.63

etheless

0.63

Activations Density 0.154%

mentions of news or information

phrases that begin with "news of" followed by significant events or occurrences

No Comments

No Known Activations