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

tta

-0.71

 Ambro

-0.71

nik

-0.70

clusions

-0.69

elman

-0.67

ffield

-0.67

umbnail

-0.67

buster

-0.66

unts

-0.66

arten

-0.64

POSITIVE LOGITS

asca

0.74

Church

0.73

ote

0.68

undown

0.67

Republic

0.67

oth

0.66

 srfAttach

0.66

ĵĺ

0.65

ighter

0.65

Past

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.