INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 licences

-0.73

 licenses

-0.68

 pals

-0.66

 leases

-0.65

 mates

-0.65

 licence

-0.65

 Maduro

-0.64

 mate

-0.64

 Mock

-0.63

 reservations

-0.62

POSITIVE LOGITS

aukee

0.76

anyahu

0.74

etsk

0.72

ablishment

0.72

gat

0.70

emer

0.70

maxwell

0.69

iger

0.69

DCS

0.69

hig

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.