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

 Tracker

-0.82

OME

-0.78

OPLE

-0.73

Tube

-0.72

Ã´

-0.67

OTOS

-0.67

 slideshow

-0.66

POSE

-0.66

asures

-0.66

ICES

-0.66

POSITIVE LOGITS

luaj

0.79

)]

0.71

 EntityItem

0.71

natural

0.65

 recip

0.64

ihu

0.63

ĳå£«

0.63

verse

0.63

jun

0.63

roe

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.