INDEX

Explanations

mentions of additional or special features in a product or content

oai_token-act-pair · gpt-3.5-turbo

references to additional content or special features

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

Negative Logits

ghai

-0.79

iard

-0.70

 Twain

-0.66

 STATES

-0.65

 Massachusetts

-0.64

 Becker

-0.62

 Grow

-0.62

 Grad

-0.62

Fram

-0.61

 Pillar

-0.61

POSITIVE LOGITS

ordinary

1.39

 extras

1.25

verted

0.96

extra

0.95

 Extras

0.92

ilogy

0.82

avorite

0.80

visor

0.79

judicial

0.79

verts

0.77

Activations Density 0.008%

mentions of additional or special features in a product or content

references to additional content or special features

No Known Activations