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

mass

-0.73

Reviewer

-0.70

 Solutions

-0.66

ndra

-0.66

Aid

-0.64

 Pier

-0.62

hin

-0.61

 basin

-0.61

damage

-0.60

 Instit

-0.59

POSITIVE LOGITS

izoph

0.87

aciously

0.76

anned

0.72

obbies

0.72

uitous

0.69

uously

0.69

oreal

0.67

 anecd

0.66

 interchangeable

0.65

inguishable

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.