INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

idays

-0.70

£

-0.65

stice

-0.65

````

-0.64

 Deity

-0.62

Impl

-0.62

½

-0.60

Provider

-0.59

uxe

-0.59

 alias

-0.58

POSITIVE LOGITS

nr

0.74

folk

0.70

pal

0.68

raged

0.67

seek

0.66

emark

0.64

pees

0.64

 Yards

0.63

ioned

0.62

rats

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.