INDEX

Explanations

phrases indicating a suggestion, recommendation, or inquiry

oai_token-act-pair · gpt-3.5-turbo

conditional phrases that pose questions or hypotheses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Eye

-0.85

ulo

-0.73

emetery

-0.72

voice

-0.71

incial

-0.71

icted

-0.70

alm

-0.67

ufact

-0.67

ocker

-0.67

âĵĺ

-0.66

POSITIVE LOGITS

 anyone

1.12

 there

1.12

 anybody

1.05

 anything

1.01

 they

0.97

fy

0.90

any

0.85

it

0.83

we

0.82

you

0.80

Activations Density 0.049%

phrases indicating a suggestion, recommendation, or inquiry

conditional phrases that pose questions or hypotheses

No Comments

No Known Activations