INDEX

Explanations

references to specific seasons of television series

oai_token-act-pair · gpt-3.5-turbo

references to television series and their seasons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

artisan

-0.85

ulic

-0.80

roleum

-0.79

ntil

-0.79

iasis

-0.75

yre

-0.75

ertodd

-0.68

ousse

-0.67

hend

-0.67

ronics

-0.66

POSITIVE LOGITS

 finale

1.30

 premiere

1.17

 episodes

1.17

 installments

1.11

 episode

1.11

isodes

1.09

 premie

1.07

 airing

1.06

 Episode

1.06

 premiered

1.02

Activations Density 0.100%

references to specific seasons of television series

references to television series and their seasons

No Comments

No Known Activations