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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Centauri

-0.73

 contag

-0.71

 reprodu

-0.70

 Indies

-0.69

 Pact

-0.67

EVE

-0.64

 plague

-0.63

 chilled

-0.62

 reproduce

-0.60

ookie

-0.60

POSITIVE LOGITS

plet

0.77

chet

0.75

kson

0.74

cks

0.72

kens

0.68

alist

0.68

ricks

0.67

ULL

0.66

âĸĳ

0.66

tti

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.