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

anna

-0.95

phe

-0.79

ona

-0.78

SON

-0.78

eryl

-0.76

âĹ¼

-0.73

rika

-0.73

rial

-0.72

oreal

-0.71

fare

-0.70

POSITIVE LOGITS

EStreamFrame

0.67

 incorpor

0.63

 railways

0.62

 unres

0.61

>.

0.60

 Shuttle

0.58

 chickens

0.58

 Downloads

0.57

 begg

0.57

 communists

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.