INDEX

Explanations

references to television shows and movie franchises

oai_token-act-pair · gpt-3.5-turbo

mentions of television shows and their seasons or related productions

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

 seiz

-0.78

 sshd

-0.68

âĸ

-0.64

ickr

-0.63

Tip

-0.62

 squats

-0.62

Geh

-0.61

 Hague

-0.61

 pian

-0.61

 smells

-0.61

POSITIVE LOGITS

 sequels

1.52

 trilogy

1.35

 sequel

1.30

quel

1.28

 reboot

1.27

ilogy

1.15

 theatrical

1.13

 installments

1.11

series

1.08

 standalone

1.08

Activations Density 0.345%

references to television shows and movie franchises

mentions of television shows and their seasons or related productions

No Comments

No Known Activations