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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

jet

-0.71

Fan

-0.66

lime

-0.64

pi

-0.64

ample

-0.63

chip

-0.62

ovan

-0.62

Ont

-0.60

quick

-0.60

 Wave

-0.60

POSITIVE LOGITS

itiz

0.84

 Randolph

0.71

ITIES

0.68

stood

0.67

 strugg

0.66

cies

0.66

 Livingston

0.65

 Fargo

0.64

Lod

0.63

ingly

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.