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

é¾įå¥ĳå£«

-0.71

 DeVos

-0.68

redd

-0.64

 lipstick

-0.63

olan

-0.62

 âĢķ

-0.61

EStream

-0.61

daq

-0.61

etsk

-0.60

REDACTED

-0.57

POSITIVE LOGITS

ace

0.79

 Berm

0.70

senal

0.69

udos

0.67

rompt

0.66

Neigh

0.66

 Links

0.66

animate

0.65

imes

0.64

tackle

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.