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

-0.87

Mub

-0.82

 mosqu

-0.74

Tec

-0.71

 Ambro

-0.71

 Saints

-0.70

emale

-0.69

rez

-0.66

 Article

-0.65

 sterling

-0.65

POSITIVE LOGITS

osa

0.78

srf

0.76

RGB

0.74

ulously

0.69

ulative

0.69

ulation

0.66

ursion

0.65

kt

0.65

 loop

0.65

olt

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.