INDEX

Explanations

No Explanations Found

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

 Values

-0.71

itudes

-0.69

äºĶ

-0.64

inders

-0.63

Ni

-0.62

renches

-0.62

izations

-0.62

ize

-0.62

depend

-0.61

 odds

-0.61

POSITIVE LOGITS

ople

0.72

sterdam

0.68

ulous

0.66

 phased

0.64

atto

0.64

getic

0.64

butt

0.64

opal

0.63

 honoured

0.63

 accomplished

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.