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

SPONSORED

-0.83

RH

-0.70

GN

-0.69

Anonymous

-0.68

Sov

-0.67

OF

-0.65

NCT

-0.65

hend

-0.64

READ

-0.64

CI

-0.64

POSITIVE LOGITS

 traveled

0.67

eri

0.66

 simulate

0.66

 strokes

0.66

 interpol

0.63

rosso

0.63

idad

0.63

 calculates

0.62

ahan

0.61

 calculations

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.