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

regon

-0.88

artisan

-0.81

taboola

-0.79

chance

-0.73

indal

-0.66

ods

-0.66

 subdivision

-0.65

rh

-0.63

throp

-0.63

}}}

-0.63

POSITIVE LOGITS

 Brav

0.80

 suits

0.72

Buc

0.67

ãĤ»

0.67

ettes

0.67

istance

0.67

Prot

0.67

ullivan

0.66

igans

0.65

Sullivan

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.