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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ription

-0.85

SHARE

-0.85

HIP

-0.80

 await

-0.80

 elig

-0.70

çīĪ

-0.69

*/(

-0.65

 awaits

-0.65

 Utilities

-0.62

 unavailable

-0.62

POSITIVE LOGITS

igmatic

0.79

zel

0.77

zee

0.73

ston

0.73

ener

0.71

zo

0.67

books

0.66

cible

0.65

metal

0.64

lights

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.