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

 Fortune

-0.75

ATK

-0.69

tis

-0.67

asters

-0.62

us

-0.60

MIT

-0.60

te

-0.59

EMP

-0.57

 grant

-0.57

 scrimmage

-0.56

POSITIVE LOGITS

angan

0.81

cohol

0.78

ãĤ®

0.78

imar

0.74

juven

0.72

arez

0.71

ikuman

0.70

yip

0.68

agi

0.68

sexual

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.