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

Negative Logits

 millenn

-0.72

ournals

-0.70

 Administ

-0.69

²¾

-0.64

 STAT

-0.61

 acknow

-0.61

 Dept

-0.61

OTAL

-0.61

 dens

-0.61

imates

-0.60

POSITIVE LOGITS

andom

0.72

cy

0.70

TTL

0.70

Yak

0.68

ishment

0.68

zel

0.67

illus

0.66

clusive

0.66

zer

0.66

witch

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.