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

owl

-0.68

 McInt

-0.67

album

-0.65

ocker

-0.65

 Ender

-0.64

ibaba

-0.64

 grading

-0.63

older

-0.62

 Equal

-0.62

iser

-0.61

POSITIVE LOGITS

ãĥ¼ãĥĨãĤ£

0.90

MODE

0.73

ãĤ¢

0.72

rawdownloadcloneembedreportprint

0.72

nces

0.71

conom

0.69

ãĤ®

0.69

wolves

0.69

ãĥīãĥ©

0.69

posed

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.