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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ufact

-0.93

 tiss

-0.84

interstitial

-0.83

vertisement

-0.77

arling

-0.73

earance

-0.73

arat

-0.72

luster

-0.72

 capacities

-0.72

ient

-0.70

POSITIVE LOGITS

OSH

0.69

Released

0.68

MPG

0.67

Way

0.66

=-=-=-=-

0.65

Rac

0.65

 Chambers

0.64

 Moses

0.63

 Overse

0.63

BBC

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.