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

kefeller

-0.75

 pale

-0.68

igl

-0.61

 millenn

-0.60

ople

-0.59

iate

-0.59

 Corridor

-0.59

agg

-0.59

elligence

-0.58

ARE

-0.58

POSITIVE LOGITS

Â·Â·

0.78

ãĥł

0.72

due

0.71

uyomi

0.70

NPR

0.67

APD

0.66

PID

0.65

orders

0.62

ubuntu

0.62

ãĤī

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.