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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

igree

-0.78

ulations

-0.76

-0.73

 Mongolia

-0.72

ioxide

-0.72

 ILCS

-0.71

iameter

-0.70

gregation

-0.70

ciating

-0.70

ceptor

-0.70

POSITIVE LOGITS

 guiName

0.72

unn

0.71

 Appeal

0.67

 architect

0.65

 usur

0.65

lein

0.65

 Feminist

0.63

FUL

0.63

 paved

0.63

 abst

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.