INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

clus

-0.70

 chunks

-0.68

ktop

-0.68

ollower

-0.65

zing

-0.61

issance

-0.59

 Lovecraft

-0.59

PART

-0.59

 tangible

-0.58

BILITIES

-0.58

POSITIVE LOGITS

istani

0.81

nesday

0.80

 Seym

0.77

spin

0.70

osate

0.70

 ********************************

0.70

enberg

0.70

yrinth

0.68

rador

0.67

Interstitial

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.