INDEX

Explanations

discussions about television show episode lengths and viewer behaviors

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

topics related to television programming and viewing behaviors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

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_post

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

etsk

-0.67

£ı

-0.67

 humility

-0.66

etically

-0.66

Merit

-0.64

 organizational

-0.63

 Gibbs

-0.62

Ĭ±

-0.62

 lender

-0.61

¬¼

-0.61

POSITIVE LOGITS

 binge

1.60

 watched

1.55

 watch

1.50

 Hulu

1.48

 watching

1.46

watching

1.41

 Watching

1.36

 WATCH

1.34

 Netflix

1.34

DVD

1.31

Activations Density 0.752%

discussions about television show episode lengths and viewer behaviors

topics related to television programming and viewing behaviors

No Comments

No Known Activations