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

mberg

-0.72

SPONSORED

-0.68

 waivers

-0.66

 preservation

-0.64

isner

-0.63

 Strongh

-0.63

 comprom

-0.63

 Tatt

-0.61

aukee

-0.61

 Percentage

-0.61

POSITIVE LOGITS

Sah

0.84

rome

0.83

Lyn

0.76

train

0.76

xx

0.73

fox

0.73

Eva

0.71

holiday

0.70

eon

0.70

Sport

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.