INDEX

Explanations

terms related to magnetic fields and magnetic technology

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ership

-0.74

semble

-0.70

cham

-0.68

FORE

-0.67

ULTS

-0.65

OTAL

-0.64

 PsyNet

-0.64

Kut

-0.63

soDeliveryDate

-0.63

POSITIVE LOGITS

 resonance

1.18

netic

1.18

 Reson

0.90

 compass

0.88

 susceptibility

0.87

 poles

0.83

 coils

0.82

 stripe

0.82

 magnetic

0.78

 monop

0.77

Activations Density 0.020%

terms related to magnetic fields and magnetic technology

No Comments

No Known Activations