INDEX

Explanations

phrases mentioning positions or titles followed by people's names

oai_token-act-pair · gpt-3.5-turbo

phrases or references to individuals or entities identified by "of" in a context that indicates ownership or association

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 expans

-0.67

obyl

-0.67

 dehuman

-0.66

 humili

-0.64

 reperto

-0.63

avi

-0.62

 exhaust

-0.59

rina

-0.59

ynes

-0.58

ories

-0.58

POSITIVE LOGITS

 Honor

0.82

 sorts

0.80

 Excellence

0.78

icial

0.77

sky

0.72

sen

0.72

 Anaheim

0.72

 course

0.71

ahu

0.69

 whom

0.68

Activations Density 0.074%

phrases mentioning positions or titles followed by people's names

phrases or references to individuals or entities identified by "of" in a context that indicates ownership or association

No Comments

No Known Activations