INDEX

Explanations

references to movies

oai_token-act-pair · gpt-3.5-turbo

mentions of films or movies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

cffff

-0.85

 prevailing

-0.66

regulated

-0.64

nyder

-0.64

Dew

-0.64

bley

-0.63

esse

-0.62

imov

-0.61

 overloaded

-0.61

lein

-0.61

POSITIVE LOGITS

film

1.01

 film

0.95

Film

0.91

 films

0.91

 Surviv

0.86

 adaptation

0.86

arium

0.83

ctors

0.83

 theaters

0.82

ography

0.82

Activations Density 0.036%

references to movies

mentions of films or movies

No Comments

No Known Activations