INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

×

-0.75

————

-0.70

--------------------

-0.65

iasis

-0.64

orno

-0.62

adoes

-0.62

ya

-0.61

otally

-0.61

 Comic

-0.60

onite

-0.60

POSITIVE LOGITS

ãĥ¯ãĥ³

0.68

FK

0.66

AAF

0.65

OTOS

0.65

 advoc

0.63

Org

0.63

erest

0.60

urgical

0.59

llular

0.59

EVA

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.