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

Feature

-0.65

feature

-0.62

MUS

-0.62

aughters

-0.62

anova

-0.61

matic

-0.61

abo

-0.61

ARY

-0.59

atural

-0.59

aments

-0.58

POSITIVE LOGITS

Magikarp

0.69

plet

0.67

imester

0.65

REDACTED

0.64

lua

0.63

Lenin

0.63

DonaldTrump

0.62

 arsen

0.62

zel

0.61

rage

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.