INDEX

Explanations

mentions of specific legislation or policy names

oai_token-act-pair · gpt-3.5-turbo

references to the entity labeled "EN" and its associated context or versions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

vati

-0.81

heng

-0.69

ffe

-0.67

illac

-0.66

 fentanyl

-0.65

cies

-0.62

FactoryReloaded

-0.62

 undone

-0.62

-0.61

 Louie

-0.61

POSITIVE LOGITS

EN

0.95

DERR

0.83

acci

0.83

vironment

0.81

terness

0.80

IG

0.80

EST

0.79

PUT

0.78

 ANGEL

0.78

TRY

0.77

Activations Density 0.009%

mentions of specific legislation or policy names

references to the entity labeled "EN" and its associated context or versions

No Comments

No Known Activations