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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SPONSORED

-0.97

adeon

-0.86

orno

-0.80

NG

-0.79

HI

-0.79

IA

-0.75

galitarian

-0.72

ESA

-0.72

ochet

-0.71

IUM

-0.70

POSITIVE LOGITS

 above

0.72

 Peng

0.71

 infall

0.71

 Weasley

0.71

 afore

0.70

 insepar

0.67

 savvy

0.67

 pursu

0.66

 inexper

0.65

worm

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.