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

 Regions

-0.78

ãĥ¼ãĥĨ

-0.78

worth

-0.74

 Nieto

-0.67

 Recall

-0.66

è¡

-0.65

©¶æ¥µ

-0.64

amac

-0.63

 Hills

-0.62

 Commands

-0.61

POSITIVE LOGITS

pter

0.78

zsche

0.70

ischer

0.68

arten

0.68

artist

0.67

chool

0.67

ript

0.65

oteric

0.65

othal

0.64

 concentrate

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.