INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ãĥ¯ãĥ³

-0.82

nown

-0.76

numbered

-0.74

 counters

-0.72

period

-0.69

canon

-0.67

eq

-0.66

wb

-0.65

quet

-0.64

perture

-0.63

POSITIVE LOGITS

enance

0.76

inav

0.68

geist

0.65

gdala

0.64

agogue

0.63

 Archdemon

0.63

orge

0.63

 shalt

0.63

 Provide

0.62

 âĢº

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.