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

 judgement

-0.68

 shif

-0.68

 yourselves

-0.67

 incap

-0.65

 crooked

-0.64

 incor

-0.63

 Triple

-0.63

 heavenly

-0.62

 Veteran

-0.62

ģ

-0.62

POSITIVE LOGITS

endor

0.91

kamp

0.77

ãĥĺ

0.76

Minecraft

0.73

yah

0.73

ãĤ¶

0.70

gren

0.70

 Tanzania

0.68

DERR

0.68

ãĥīãĥ©

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.