INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ãģĹ

-0.66

ĸļ

-0.63

 redes

-0.61

 Reboot

-0.59

SHIP

-0.58

ATH

-0.58

ISSION

-0.57

enhagen

-0.57

encing

-0.56

INTON

-0.56

POSITIVE LOGITS

rooms

0.77

odor

0.73

sold

0.72

lav

0.71

zers

0.71

holder

0.71

gal

0.71

zer

0.69

hor

0.68

room

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.