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

 ante

-0.73

idon

-0.72

 Uriel

-0.69

llan

-0.67

Fall

-0.65

 Shall

-0.64

lain

-0.63

 Milton

-0.63

xus

-0.62

 Payne

-0.59

POSITIVE LOGITS

ofi

0.83

Ô

0.78

fusc

0.71

CBI

0.71

orneys

0.70

Trade

0.70

iminary

0.67

opy

0.67

TERN

0.66

Ĥ

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.