INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ofi

-0.75

 membr

-0.63

 breathe

-0.62

piece

-0.61

 Giles

-0.60

gou

-0.59

ape

-0.59

 enlarg

-0.59

bringer

-0.58

èĢħ

-0.58

POSITIVE LOGITS

weekly

0.76

Redditor

0.75

Alert

0.73

undown

0.73

VERTISEMENT

0.70

ARDS

0.67

Rx

0.63

Wallet

0.63

rams

0.63

warm

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.