INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 buff

-0.67

 bundles

-0.63

assetsadobe

-0.61

SY

-0.59

raits

-0.57

 tant

-0.57

 Saddam

-0.57

 accomplished

-0.56

 doct

-0.56

uz

-0.56

POSITIVE LOGITS

iated

0.79

perty

0.79

oland

0.74

entric

0.72

pard

0.71

ledged

0.70

ICO

0.69

mares

0.67

staking

0.66

Cook

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.