INDEX

Explanations

mentions of the word "movies"

oai_token-act-pair · gpt-3.5-turbo

references to movies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Pwr

-0.71

rition

-0.70

staff

-0.69

holder

-0.68

 Brewer

-0.66

mouth

-0.65

ANGE

-0.65

aith

-0.64

entry

-0.63

arity

-0.62

POSITIVE LOGITS

 movies

3.63

 films

2.89

 Movies

2.84

 movie

2.25

ovies

2.14

 cinema

2.12

 documentaries

2.10

movie

2.08

 DVDs

1.99

 Films

1.94

Activations Density 0.030%

mentions of the word "movies"

references to movies

No Comments

No Known Activations