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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

den

-0.78

pot

-0.76

pid

-0.76

powered

-0.75

conn

-0.74

priced

-0.68

Pag

-0.68

painted

-0.66

equipped

-0.66

 Torment

-0.63

POSITIVE LOGITS

zos

0.74

 Thumbnails

0.73

``

0.72

ãĤĵ

0.67

IDS

0.65

 snipp

0.65

ILLE

0.63

Recomm

0.63

 Collection

0.62

rouse

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.