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

 Jail

-0.66

itely

-0.64

COP

-0.63

 jail

-0.63

ÄŁ

-0.62

erm

-0.61

 somew

-0.59

iors

-0.59

olean

-0.59

iliar

-0.58

POSITIVE LOGITS

ifax

0.75

MSN

0.75

 Lumpur

0.75

INGTON

0.71

ADRA

0.71

rax

0.69

 Rasmussen

0.68

 totality

0.67

NET

0.67

BuyableInstoreAndOnline

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.