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

 overc

-0.77

clair

-0.75

apo

-0.71

anche

-0.67

pas

-0.65

iann

-0.64

ural

-0.64

MG

-0.64

oak

-0.64

uart

-0.63

POSITIVE LOGITS

 Valhalla

0.83

 Goth

0.69

 Stard

0.68

hess

0.67

 Estonia

0.66

ihara

0.65

 Traff

0.63

 Robots

0.63

DAQ

0.62

 Soviets

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.