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

ª

-0.74

abulary

-0.69

 invari

-0.67

umbers

-0.67

âĢ¦âĢ¦âĢ¦âĢ¦

-0.66

lier

-0.65

Ã¢

-0.64

anton

-0.64

lihood

-0.63

 fare

-0.60

POSITIVE LOGITS

ãĤ¨ãĥ«

0.74

 Prototype

0.72

ãĥŃ

0.70

ahime

0.70

ãĤ½

0.66

ocalyptic

0.65

endiary

0.63

rys

0.63

 Pyro

0.62

 Adin

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.