INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

dinand

-0.76

ongyang

-0.72

uyomi

-0.72

akening

-0.72

ogun

-0.72

othal

-0.71

iterranean

-0.71

ulkan

-0.70

ithering

-0.68

farious

-0.67

POSITIVE LOGITS

 Funk

0.66

WT

0.66

 Byrd

0.66

camp

0.65

 Clara

0.65

DB

0.65

stra

0.64

cap

0.64

 Gibbs

0.63

 Span

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.