INDEX

Explanations

phrases related to the act of being viewed or perceived

oai_token-act-pair · gpt-3.5-turbo

references to perceptions of being analyzed or judged

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ammy

-0.71

ante

-0.69

ells

-0.69

angan

-0.67

dry

-0.66

 purse

-0.65

cake

-0.65

airs

-0.64

ayn

-0.64

broken

-0.64

POSITIVE LOGITS

ById

1.03

 viewed

0.89

 views

0.87

 guiActiveUn

0.86

 "$:/

0.85

ouble

0.76

CLASSIFIED

0.76

¿½

0.76

ARGET

0.73

DragonMagazine

0.72

Activations Density 0.006%

phrases related to the act of being viewed or perceived

references to perceptions of being analyzed or judged

No Comments

No Known Activations