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

pri

-0.69

Whe

-0.68

onew

-0.67

 bother

-0.64

flows

-0.62

 interpret

-0.61

isively

-0.60

jud

-0.60

oppy

-0.60

POSITIVE LOGITS

[|

0.78

senal

0.77

kamp

0.75

ãĥŁ

0.73

encers

0.70

encer

0.70

Ö¼

0.68

yrinth

0.67

rentices

0.65

ierrez

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.