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

Negative Logits

 Rash

-0.73

ãģĨ

-0.71

CPU

-0.68

SourceFile

-0.68

Ey

-0.68

 strokes

-0.64

³

-0.63

puff

-0.60

cpu

-0.59

going

-0.58

POSITIVE LOGITS

arcity

0.93

isner

0.87

ibaba

0.82

apor

0.81

issance

0.78

etsk

0.78

iphate

0.78

idia

0.77

edia

0.76

issan

0.76

Activations Density 0.000%

No Known Activations

This feature has no known activations.