INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

camp

-0.86

shed

-0.66

 seiz

-0.64

 thor

-0.63

elsen

-0.61

 thirds

-0.61

 Airl

-0.60

ready

-0.58

ledged

-0.58

lyn

-0.58

POSITIVE LOGITS

escription

0.73

 governing

0.73

ãĤ¢ãĥ«

0.70

ortium

0.68

waukee

0.67

AMI

0.66

erity

0.66

soType

0.63

GBT

0.63

 Franch

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.