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

ngth

-0.68

yout

-0.67

trop

-0.64

 caut

-0.63

paren

-0.63

 dilig

-0.62

 myster

-0.61

 warranted

-0.61

 squee

-0.59

 enqu

-0.59

POSITIVE LOGITS

illac

0.71

enum

0.71

arine

0.70

GES

0.69

ILE

0.68

umped

0.68

iled

0.67

HM

0.67

uca

0.65

lass

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.