INDEX

Explanations

mentions of specific communities or community-related locations

oai_token-act-pair · gpt-3.5-turbo

mentions of community-related institutions and locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

inventoryQuantity

-0.77

 toget

-0.76

 tread

-0.76

 anonymity

-0.75

theless

-0.74

iod

-0.72

 fingerprints

-0.70

Downloadha

-0.70

 aven

-0.68

 keyword

-0.68

POSITIVE LOGITS

 Center

1.58

 Complex

1.52

 Facility

1.52

 Resort

1.45

 School

1.44

 Centre

1.43

 Clinic

1.43

 Station

1.41

 Club

1.38

 Hospital

1.38

Activations Density 0.212%

mentions of specific communities or community-related locations

mentions of community-related institutions and locations

No Comments

No Known Activations