INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

nir

-0.70

 condos

-0.70

 blades

-0.67

 safest

-0.66

 happiest

-0.66

ãĥĥãĥĪ

-0.64

 pads

-0.64

 platinum

-0.64

 erection

-0.63

 deserts

-0.61

POSITIVE LOGITS

zek

0.76

oz

0.74

quished

0.73

avan

0.73

avis

0.72

RIP

0.72

eret

0.71

ket

0.69

asar

0.68

igor

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.