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

""

-0.73

cf

-0.67

cr

-0.66

bright

-0.64

"""

-0.64

bud

-0.63

tec

-0.63

Face

-0.63

Doc

-0.63

double

-0.62

POSITIVE LOGITS

 redes

0.89

 adolesc

0.77

 conduc

0.77

 lett

0.73

nces

0.71

 Citiz

0.69

 Poverty

0.67

 horizont

0.67

OPLE

0.67

©¶æ

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.