INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ilater

-0.76

 Rhod

-0.74

izo

-0.72

 medd

-0.69

 vigorously

-0.68

 trove

-0.67

ailable

-0.66

 tuple

-0.64

glers

-0.63

POSITIVE LOGITS

borough

0.78

christ

0.76

ev

0.76

omics

0.73

church

0.72

bones

0.71

eco

0.71

soType

0.67

 Koen

0.66

aut

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.