INDEX

Explanations

mentions of TV shows, specifically focusing on the concept of a show and its production aspects

oai_token-act-pair · gpt-3.5-turbo

references to television shows

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

 solitary

-0.68

 Aval

-0.62

 Pastebin

-0.60

 calcul

-0.60

 Hurricanes

-0.59

illet

-0.59

 agon

-0.57

 Allaah

-0.56

 bilingual

-0.55

 saline

-0.55

POSITIVE LOGITS

biz

1.83

runners

1.57

manship

1.54

runner

1.53

downs

1.44

case

1.33

room

1.31

rooms

1.27

cases

1.23

alter

1.16

Activations Density 0.056%

mentions of TV shows, specifically focusing on the concept of a show and its production aspects

references to television shows

No Comments

No Known Activations