INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

quickShipAvailable

-0.72

âĸ¬âĸ¬

-0.63

Nir

-0.63

 Autumn

-0.63

 Vern

-0.62

nces

-0.61

gravity

-0.61

 karma

-0.61

Vert

-0.60

 Grape

-0.60

POSITIVE LOGITS

raf

0.85

IS

0.76

iddling

0.75

atel

0.75

ulner

0.72

MRI

0.72

zsche

0.71

bid

0.71

 Concern

0.70

deb

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.