INDEX

Explanations

mentions of movies

oai_token-act-pair · gpt-3.5-turbo

references to 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

èª

-0.92

uous

-0.83

colo

-0.78

ifts

-0.74

å£

-0.71

uously

-0.71

oise

-0.70

IBLE

-0.69

ought

-0.69

ä¹ĭ

-0.68

POSITIVE LOGITS

 Mayhem

1.09

 Madness

0.95

 Matters

0.85

 Mania

0.83

 Improvement

0.80

 Definition

0.79

 Interest

0.79

 Disorders

0.78

 Massacre

0.76

achus

0.76

Activations Density 0.076%

mentions of movies

references to movies

No Comments

No Known Activations