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

iege

-0.71

 Crus

-0.68

 Bohem

-0.68

 Wester

-0.67

IBLE

-0.67

 Vide

-0.66

 Chall

-0.66

 Peel

-0.64

gard

-0.64

 Crusader

-0.64

POSITIVE LOGITS

 fooled

0.77

 screwed

0.69

Zip

0.67

onents

0.65

 hooked

0.64

umpy

0.61

 precincts

0.61

nces

0.60

pta

0.60

aster

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.