INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ndra

-0.72

Roe

-0.69

zsche

-0.69

vi

-0.69

 Recomm

-0.68

 Hendricks

-0.67

 Elixir

-0.66

velt

-0.66

Rx

-0.65

chwitz

-0.64

POSITIVE LOGITS

rained

0.74

 istg

0.68

Latin

0.68

cpu

0.65

PsyNetMessage

0.63

united

0.62

 whim

0.61

pun

0.60

lat

0.59

oca

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.