INDEX

Explanations

organizations or groups named "Alliance" with varying levels of emphasis

oai_token-act-pair · gpt-3.5-turbo

references to an organization or coalition named "Alliance."

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

agar

-0.80

agate

-0.78

amera

-0.75

gone

-0.74

asus

-0.72

things

-0.71

ochemical

-0.70

oses

-0.68

animate

-0.68

artifacts

-0.68

POSITIVE LOGITS

 Alliance

1.04

 FranÃ§

0.89

naire

0.75

venture

0.70

 Provision

0.69

ison

0.68

 Partnership

0.68

 Network

0.68

 Coalition

0.68

 alliance

0.68

Activations Density 0.012%

organizations or groups named "Alliance" with varying levels of emphasis

references to an organization or coalition named "Alliance."

No Comments

No Known Activations