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

SPONSORED

-1.04

CLASSIFIED

-0.74

Rect

-0.71

 Territories

-0.69

FactoryReloaded

-0.67

FontSize

-0.65

inct

-0.65

@#

-0.64

#$

-0.64

POSITIVE LOGITS

ikarp

0.98

aeus

0.68

glers

0.67

icipated

0.67

ibaba

0.64

st

0.63

ishers

0.62

 favorites

0.61

otos

0.61

resa

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.