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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 testim

-0.73

 quot

-0.71

»Ĵ

-0.68

etheless

-0.67

ãĤ´ãĥ³

-0.63

 savage

-0.62

 unsub

-0.61

 somet

-0.59

 whore

-0.59

ãĤ±

-0.59

POSITIVE LOGITS

PLIED

0.84

 Cars

0.81

 Yard

0.72

flies

0.71

"},"

0.68

 Cannot

0.65

orsi

0.64

conom

0.64

 DRAG

0.63

 transit

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.