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

TL

-0.75

asus

-0.73

AU

-0.69

AMD

-0.69

WI

-0.66

DeL

-0.65

GL

-0.64

Alien

-0.64

SG

-0.64

OTHER

-0.63

POSITIVE LOGITS

plings

0.86

esty

0.85

ges

0.78

itud

0.73

icrobial

0.70

 Relief

0.70

 toile

0.69

ignt

0.68

lied

0.67

amina

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.