INDEX

Explanations

adjectives related to competency and completeness

oai_token-act-pair · gpt-3.5-turbo

attributes related to quality or competence

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

Not in Any Lists

No Comments

Negative Logits

KEY

-0.71

ONEY

-0.70

eeks

-0.70

aucus

-0.69

LAN

-0.66

erves

-0.65

ften

-0.65

osponsors

-0.63

eper

-0.63

VPN

-0.63

POSITIVE LOGITS

tenance

1.03

 survival

0.67

osterone

0.65

 subsequ

0.64

acters

0.64

err

0.63

 offspring

0.62

ness

0.62

urance

0.62

 sibling

0.62

Activations Density 0.151%

adjectives related to competency and completeness

attributes related to quality or competence

No Comments

No Known Activations