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

genic

-0.68

ournals

-0.64

ollah

-0.62

markets

-0.61

auc

-0.60

ulz

-0.59

rossover

-0.59

 Airl

-0.58

fam

-0.58

bek

-0.58

POSITIVE LOGITS

atis

0.86

 Wage

0.72

MENT

0.68

CLASS

0.68

mental

0.67

tale

0.66

tons

0.66

SpaceEngineers

0.66

sbm

0.64

ton

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.