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

rone

-0.74

iverse

-0.69

iyah

-0.67

actionDate

-0.66

ola

-0.65

oil

-0.65

URRENT

-0.64

[/

-0.64

olin

-0.63

cation

-0.62

POSITIVE LOGITS

 Berks

0.82

hma

0.76

 reins

0.70

rily

0.69

rosse

0.68

 levers

0.64

 lever

0.64

 Rober

0.63

imore

0.63

 penny

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.