INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Klu

-0.79

Gry

-0.71

ullah

-0.70

 Venezuel

-0.68

Fi

-0.68

 Russians

-0.68

 Corpus

-0.67

Oro

-0.67

Mub

-0.66

gib

-0.65

POSITIVE LOGITS

ourney

0.75

model

0.75

icer

0.73

elling

0.73

teenth

0.71

ients

0.70

amping

0.70

isations

0.70

seys

0.70

onga

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.