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

retty

-0.82

ALSE

-0.82

Cand

-0.72

actionDate

-0.72

irtual

-0.71

RIPT

-0.71

 adequ

-0.70

incerity

-0.70

oters

-0.70

senal

-0.69

POSITIVE LOGITS

 Watt

0.85

 Cerberus

0.81

worth

0.71

 Perkins

0.68

 Grimm

0.67

Von

0.64

 Leviathan

0.63

 Freed

0.63

Â·Â·

0.62

 Tyrann

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.