INDEX

Explanations

titles of movies, TV shows, and video games

oai_token-act-pair · gpt-3.5-turbo

references to movies and their sequels or related media

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

reek

-0.83

 insulated

-0.70

informed

-0.69

arse

-0.68

fortable

-0.68

 knowledgeable

-0.68

 econom

-0.67

 territ

-0.67

 gull

-0.64

isk

-0.64

POSITIVE LOGITS

 sequels

0.98

Transform

0.93

 cinematic

0.89

 soundtrack

0.89

 Invasion

0.86

 Kinect

0.86

netflix

0.86

 finale

0.83

Released

0.83

ilogy

0.83

Activations Density 0.625%

titles of movies, TV shows, and video games

references to movies and their sequels or related media

No Comments

No Known Activations