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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Cosponsors

-0.75

OT

-0.71

Ring

-0.68

GROUP

-0.68

OTS

-0.67

CHQ

-0.67

ordinate

-0.67

OC

-0.65

ACTION

-0.64

lled

-0.63

POSITIVE LOGITS

 Wast

0.82

opian

0.74

 Veter

0.73

emon

0.72

Ëľ

0.69

ean

0.69

estine

0.68

Cum

0.67

akeru

0.67

 Yuan

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.