INDEX

Explanations

mentions of cinema or related activities

oai_token-act-pair · gpt-3.5-turbo

references to cinema and film-related contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lies

-0.80

nesses

-0.76

essee

-0.74

tnc

-0.72

laus

-0.72

ional

-0.71

lers

-0.71

lying

-0.70

bek

-0.66

opher

-0.65

POSITIVE LOGITS

goers

0.92

ovies

0.91

 cinema

0.82

 marqu

0.80

 cinem

0.80

 theaters

0.80

 premie

0.79

 Noir

0.77

 Cinema

0.77

 studios

0.75

Activations Density 0.021%

mentions of cinema or related activities

references to cinema and film-related contexts

No Comments

No Known Activations