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

REAM

-0.68

 exponent

-0.62

 laure

-0.61

 dens

-0.61

 bands

-0.61

PASS

-0.61

 vividly

-0.60

 regimes

-0.60

ategor

-0.60

resso

-0.59

POSITIVE LOGITS

ahime

0.87

=-=-=-=-

0.79

court

0.78

thumbnails

0.78

cour

0.74

Gh

0.72

Sham

0.72

gdala

0.72

Hug

0.71

å¥³

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.