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

asonic

-0.74

payers

-0.71

ãĤ±

-0.68

BALL

-0.67

peat

-0.67

apps

-0.64

 oven

-0.64

beer

-0.63

acher

-0.63

ModLoader

-0.62

POSITIVE LOGITS

ĺħ

0.86

inct

0.67

ionics

0.67

edia

0.66

Ved

0.66

enza

0.64

 superficial

0.63

stract

0.63

 rational

0.60

lass

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.