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

[/

-0.70

scenes

-0.69

 tabl

-0.67

 desper

-0.66

 bosses

-0.66

clips

-0.65

gotten

-0.65

sheets

-0.64

uca

-0.63

 habitual

-0.63

POSITIVE LOGITS

ONSORED

0.87

ãĤ¡

0.81

ãĤ´ãĥ³

0.74

 Doors

0.71

 Mosque

0.69

Pwr

0.67

 Ware

0.65

URI

0.65

é¾

0.65

 Meal

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.