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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 controlled

-0.75

 deductions

-0.69

 marrow

-0.69

 appraisal

-0.69

 realities

-0.65

 deduction

-0.63

 tempor

-0.61

 cultures

-0.59

 Leban

-0.59

 conditioned

-0.59

POSITIVE LOGITS

nm

0.88

egg

0.77

uously

0.76

argon

0.76

iris

0.75

helm

0.75

rx

0.75

ruck

0.74

flush

0.71

bush

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.