INDEX

Explanations

mentions or discussions of movie premieres

oai_token-act-pair · gpt-3.5-turbo

references to the term "premiere" in relation to shows or films

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

nuts

-0.81

sie

-0.69

sle

-0.69

agh

-0.67

ila

-0.66

akia

-0.65

surface

-0.65

ansom

-0.64

ateg

-0.64

arms

-0.64

POSITIVE LOGITS

 premiere

1.07

 screening

0.95

 premie

0.94

 Premiere

0.94

 airing

0.93

 screenings

0.90

 reel

0.83

 episode

0.81

APR

0.80

 simultane

0.80

Activations Density 0.010%

mentions or discussions of movie premieres

references to the term "premiere" in relation to shows or films

No Comments

No Known Activations