INDEX

Explanations

locations within a building, particularly rooms like the back seat, teacher's desk, laundry, dressing room, and interrogation room

oai_token-act-pair · gpt-3.5-turbo

words related to physical locations or settings

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

urrencies

-0.69

 Targ

-0.65

 Composite

-0.64

 Worldwide

-0.64

 Flavor

-0.63

 divor

-0.63

 Syndrome

-0.61

 winters

-0.61

 Skies

-0.61

 semantics

-0.61

POSITIVE LOGITS

 booth

1.43

 room

1.35

room

1.32

 desk

1.23

 station

1.16

 compartment

1.15

 table

1.11

 locker

1.10

 aisle

1.10

 doorway

1.10

Activations Density 0.325%

locations within a building, particularly rooms like the back seat, teacher's desk, laundry, dressing room, and interrogation room

words related to physical locations or settings

No Comments

No Known Activations