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

estern

-0.74

 DRAG

-0.72

 taxpayer

-0.66

iHUD

-0.65

 offence

-0.64

 proceeds

-0.62

è£ħ

-0.62

SUR

-0.61

 dearly

-0.61

stood

-0.60

POSITIVE LOGITS

ority

0.84

 Fiction

0.71

olia

0.68

 Romance

0.67

 Blossom

0.67

ync

0.66

wcsstore

0.65

pers

0.65

isure

0.64

dylib

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.