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

bars

-0.70

anasia

-0.68

 launchers

-0.67

acan

-0.65

rop

-0.65

ictions

-0.65

rises

-0.63

roc

-0.62

ERA

-0.61

apo

-0.61

POSITIVE LOGITS

senal

0.76

 tradem

0.74

ß

0.67

 Lisbon

0.66

ãĥı

0.66

ãĥ´ãĤ¡

0.66

 Polk

0.65

ModLoader

0.65

ãĤ¼

0.65

 Leban

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.