INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

\\\\\\\\

-0.78

ARB

-0.76

anan

-0.76

yn

-0.75

ola

-0.74

rat

-0.74

iv

-0.74

Pand

-0.73

onom

-0.71

è¦ļéĨĴ

-0.70

POSITIVE LOGITS

etheless

0.87

jong

0.84

merce

0.77

ellery

0.70

 drunken

0.68

lihood

0.68

 Ronaldo

0.68

 Fernand

0.67

 blot

0.67

 Notting

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.