INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

utations

-0.74

 thrive

-0.70

 Flip

-0.66

 someday

-0.64

ĺħ

-0.64

FC

-0.62

acy

-0.62

 boosters

-0.62

essor

-0.61

 Stacy

-0.61

POSITIVE LOGITS

illac

0.75

rawdownloadcloneembedreportprint

0.68

cat

0.68

GES

0.67

 cyan

0.67

iasco

0.65

é»Ĵ

0.63

oxide

0.62

bowl

0.62

heart

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.