INDEX

Explanations

image captions

oai_token-act-pair · gpt-3.5-turbo

references to image captions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

nex

-0.71

cius

-0.70

ASE

-0.69

 delinquent

-0.67

ISA

-0.66

CI

-0.66

lli

-0.64

 hereby

-0.64

hibited

-0.63

hibit

-0.62

POSITIVE LOGITS

 caption

0.90

acters

0.90

 Annotations

0.85

 Murd

0.80

 IMAGES

0.79

 Transcript

0.75

 Saga

0.75

PLIED

0.74

 Scenes

0.72

 Provided

0.71

Activations Density 0.007%

image captions

references to image captions

No Comments

No Known Activations