INDEX

Explanations

proper nouns related to movies and film industry

oai_token-act-pair · gpt-3.5-turbo

film titles and notable works in cinema

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

 podium

-0.77

PSU

-0.77

 benches

-0.71

 exch

-0.70

 constitu

-0.67

 textile

-0.66

 knowledgeable

-0.64

 porous

-0.63

 regist

-0.63

 leve

-0.62

POSITIVE LOGITS

 Girl

0.93

 Fever

0.93

 Days

0.93

 Goodbye

0.92

 Tomorrow

0.91

 Cowboy

0.90

 Deadly

0.89

 Dreams

0.87

 Falling

0.87

 Girls

0.87

Activations Density 0.416%

proper nouns related to movies and film industry

film titles and notable works in cinema

No Comments

No Known Activations