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

idon

-0.85

 sidx

-0.73

ÂłÂłÂłÂłÂłÂłÂłÂł

-0.70

 panc

-0.69

 Blumenthal

-0.67

Wak

-0.64

Nak

-0.64

ãĥ³ãĤ¸

-0.63

 Cobra

-0.62

Rak

-0.62

POSITIVE LOGITS

RL

0.73

algia

0.71

folk

0.69

 Fathers

0.68

iversal

0.68

rooms

0.63

birth

0.62

xx

0.62

 Loll

0.62

resses

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.