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

 TAMADRA

-0.88

 Lumpur

-0.77

 Peaks

-0.73

çīĪ

-0.73

 Mahjong

-0.72

ãĥ¥

-0.68

etooth

-0.67

 mosqu

-0.66

 "$:/

-0.66

FTWARE

-0.66

POSITIVE LOGITS

ciating

0.85

ann

0.83

ussion

0.81

elect

0.74

massive

0.74

intend

0.74

negative

0.73

sense

0.71

closed

0.71

merce

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.