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

Īè

-0.80

eryl

-0.68

yah

-0.64

 illeg

-0.64

":[{"

-0.64

 Latinos

-0.63

 Bran

-0.62

itled

-0.62

 Hughes

-0.62

jan

-0.61

POSITIVE LOGITS

 Nanto

0.75

dimension

0.75

 perspect

0.66

.ï¿½

0.64

dim

0.64

 Enhance

0.63

ulative

0.63

 optimization

0.62

esville

0.62

space

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.