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

arc

-0.71

rams

-0.70

Â£

-0.70

Tx

-0.69

unt

-0.64

low

-0.64

bl

-0.63

ATT

-0.63

alin

-0.63

Cat

-0.62

POSITIVE LOGITS

 waters

1.18

chio

0.77

ãĥ¼ãĥĨ

0.76

ibrary

0.76

ĵĺ

0.75

ãĥ¼ãĥĨãĤ£

0.72

 obser

0.71

£ı

0.70

 disg

0.69

 simul

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.