INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Ys

-0.69

CN

-0.65

MOR

-0.63

·

-0.62

/_

-0.61

 Morty

-0.60

SHARE

-0.59

MY

-0.58

 Sawyer

-0.57

 JUST

-0.57

POSITIVE LOGITS

cair

0.88

oaded

0.75

cloth

0.74

cry

0.74

uminium

0.73

haw

0.71

antry

0.70

ve

0.67

ancers

0.66

maxwell

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.