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

ensions

-0.85

ashtra

-0.76

arrow

-0.70

ascript

-0.68

 dwarves

-0.66

galitarian

-0.65

Bound

-0.64

aucuses

-0.63

jriwal

-0.62

utation

-0.62

POSITIVE LOGITS

 Giov

0.80

nil

0.74

inea

0.73

Hue

0.67

wagon

0.65

 Hermes

0.65

 Hutch

0.64

icago

0.64

wagen

0.63

.............

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.