INDEX

Explanations

No Explanations Found

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

 Dating

-0.72

eur

-0.69

 Incredible

-0.66

 Liqu

-0.64

 Racing

-0.63

 Indust

-0.63

 Inquis

-0.62

 Planning

-0.61

 worlds

-0.61

LR

-0.61

POSITIVE LOGITS

ouls

0.80

ourced

0.78

boro

0.78

oup

0.77

uggest

0.77

urses

0.77

ideshow

0.76

akens

0.73

letcher

0.73

baugh

0.73

Activations Density 0.000%

No Known Activations

This feature has no known activations.