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

choes

-0.70

ixty

-0.67

Connection

-0.66

friends

-0.65

ĳå£«

-0.64

ãĤ¼

-0.64

eno

-0.63

ixt

-0.63

½

-0.62

Compan

-0.61

POSITIVE LOGITS

IOR

0.71

Production

0.70

igslist

0.68

armac

0.67

terday

0.63

 immobil

0.61

ARC

0.60

 SOLD

0.60

}"

0.59

theless

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.