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

hower

-0.80

icum

-0.77

akia

-0.74

thritis

-0.71

Interstitial

-0.69

aceous

-0.68

rum

-0.68

gerald

-0.68

tower

-0.68

ograph

-0.67

POSITIVE LOGITS

Hel

0.71

 source

0.66

Teg

0.63

 Outs

0.63

Ut

0.61

 Damian

0.60

 AUTHOR

0.60

 Rhino

0.60

Tools

0.59

bol

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.