INDEX

Explanations

word related to resolutions, decisions, and actions taken by groups or councils

oai_token-act-pair · gpt-3.5-turbo

references to formal agreements or legislative measures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

elong

-0.72

rium

-0.72

eteria

-0.70

OTAL

-0.69

spring

-0.67

FactoryReloaded

-0.66

odge

-0.65

————

-0.65

axter

-0.64

enna

-0.64

POSITIVE LOGITS

 authorizing

1.07

 condemning

0.98

 resolution

0.95

 Resolution

0.92

 resolutions

0.88

olution

0.77

 approving

0.75

 passed

0.75

 resolving

0.71

 denouncing

0.71

Activations Density 0.038%

word related to resolutions, decisions, and actions taken by groups or councils

references to formal agreements or legislative measures

No Comments

No Known Activations