INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

patrick

-0.82

itionally

-0.72

ishly

-0.67

arded

-0.64

viously

-0.64

figure

-0.63

fy

-0.62

ably

-0.62

 Shiny

-0.61

iod

-0.61

POSITIVE LOGITS

ISM

0.89

âĸ¬

0.78

DEP

0.75

Ã±o

0.73

 Calories

0.72

 Lives

0.69

 Pieces

0.69

 Canad

0.67

 Authorization

0.66

 Divide

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.