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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

vae

-0.89

emen

-0.71

stroke

-0.70

use

-0.68

Hun

-0.68

YP

-0.68

ARE

-0.67

ï¸ı

-0.67

hs

-0.65

cale

-0.65

POSITIVE LOGITS

ãĥİ

0.70

 stricken

0.67

ernaut

0.66

 tack

0.65

 prosec

0.65

ãĤ¤ãĥĪ

0.65

ãĤ´ãĥ³

0.63

 pronoun

0.62

 quizz

0.62

ä¹ĭ

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.