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

 Hawth

-0.74

ington

-0.72

Ast

-0.67

IST

-0.65

Xin

-0.62

 Dmit

-0.61

 Shake

-0.61

conn

-0.61

 Garland

-0.61

 Hayes

-0.60

POSITIVE LOGITS

mble

0.98

xual

0.94

rollers

0.70

requ

0.69

anwhile

0.68

roller

0.67

ntil

0.66

catentry

0.66

ascript

0.66

alion

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.