INDEX

Explanations

references to groups or organizations with the term "Alliance" in their name

oai_token-act-pair · gpt-3.5-turbo

references to various organizations or alliances

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

agate

-0.78

ochemical

-0.77

orders

-0.76

things

-0.74

gone

-0.72

aptic

-0.72

agar

-0.70

oses

-0.70

asus

-0.69

ebin

-0.68

POSITIVE LOGITS

 Alliance

0.99

 FranÃ§

0.98

 alliance

0.73

naire

0.72

iance

0.70

 Partnership

0.70

ande

0.70

venture

0.68

 bloc

0.67

 partner

0.67

Activations Density 0.017%

references to groups or organizations with the term "Alliance" in their name

references to various organizations or alliances

No Comments

No Known Activations