INDEX

Explanations

locations within specific neighborhoods or districts

oai_token-act-pair · gpt-3.5-turbo

references to specific neighborhoods or regions

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

 invitations

-0.73

Americans

-0.67

Reloaded

-0.66

TPP

-0.64

HMS

-0.64

 Metatron

-0.61

 Israelis

-0.60

 physicists

-0.60

etheus

-0.60

 Canaver

-0.60

POSITIVE LOGITS

 vicinity

1.61

 neighborhood

1.57

 suburb

1.57

 neighbourhood

1.55

 district

1.50

 area

1.50

 suburbs

1.49

 neighborhoods

1.39

 countryside

1.34

 districts

1.33

Activations Density 0.445%

locations within specific neighborhoods or districts

references to specific neighborhoods or regions

No Comments

No Known Activations