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

 McGee

-0.79

imov

-0.79

weights

-0.72

Sov

-0.70

qi

-0.66

verages

-0.65

Äĩ

-0.65

forces

-0.64

 Gork

-0.63

ppings

-0.63

POSITIVE LOGITS

 payday

0.69

 dock

0.64

inary

0.60

osuke

0.60

 luxury

0.59

autical

0.57

prising

0.57

 preschool

0.57

ilee

0.57

MAS

0.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.