INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

etheless

-0.83

 Entered

-0.77

 taught

-0.66

 predicting

-0.65

ÃŃs

-0.65

aspberry

-0.65

 railing

-0.64

nels

-0.64

aders

-0.64

ansen

-0.64

POSITIVE LOGITS

 Shen

0.67

 execut

0.66

 Priv

0.65

 Flowers

0.65

Sop

0.64

oth

0.63

 Harding

0.62

HIP

0.62

 Hyde

0.61

Pie

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.