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

 compr

-0.85

zbollah

-0.80

orsi

-0.78

Downloadha

-0.74

senal

-0.71

 enthus

-0.70

jriwal

-0.69

clerosis

-0.67

Random

-0.66

anyahu

-0.65

POSITIVE LOGITS

EC

0.75

earth

0.73

EF

0.73

icol

0.69

egu

0.67

ature

0.66

strom

0.65

ellen

0.65

qu

0.64

orge

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.