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

ounter

-0.72

resses

-0.70

wang

-0.66

 Bulg

-0.64

utive

-0.62

 discounts

-0.62

address

-0.61

 dated

-0.61

 vested

-0.61

 chop

-0.61

POSITIVE LOGITS

SPONSORED

0.88

 Citiz

0.81

mberg

0.71

TPPStreamerBot

0.71

oresc

0.68

TeX

0.67

ACA

0.67

icia

0.66

 denomin

0.65

 Analy

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.