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

 wagon

-0.74

 dissatisf

-0.73

ãĥ¼ãĥĨãĤ£

-0.67

çİĭ

-0.65

fir

-0.63

 revol

-0.63

 comprom

-0.63

 discont

-0.63

 ende

-0.62

è¦ļéĨĴ

-0.62

POSITIVE LOGITS

 Slug

0.77

actionDate

0.76

ven

0.72

atel

0.68

 Rapt

0.68

Lad

0.66

uel

0.65

otten

0.65

iless

0.64

mone

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.