homebrewnlp

Clashluke's group workspace

Group: embedding-gradient-shrink

4858

4854

331

1-4

of 4

Timestamps visible

2022-12-31 20:26:38

[ 18913/262144] Loss: 0.802 - Accuracy: 0.754 | LearningRate: 0.00876 | StepTime: 49.830479s - Rate: 232,083.8 Tokens/s

2022-12-31 20:27:28

[ 18917/262144] Loss: 0.803 - Accuracy: 0.754 | LearningRate: 0.00876 | StepTime: 49.824919s - Rate: 231,997.6 Tokens/s

2022-12-31 20:28:18

[ 18921/262144] Loss: 0.802 - Accuracy: 0.755 | LearningRate: 0.00876 | StepTime: 49.821276s - Rate: 231,911.5 Tokens/s

2022-12-31 20:29:08

[ 18925/262144] Loss: 0.809 - Accuracy: 0.752 | LearningRate: 0.00876 | StepTime: 49.822121s - Rate: 231,825.5 Tokens/s

2022-12-31 20:29:58

[ 18929/262144] Loss: 0.807 - Accuracy: 0.752 | LearningRate: 0.00876 | StepTime: 49.817723s - Rate: 231,739.6 Tokens/s

2022-12-31 20:30:48

[ 18933/262144] Loss: 0.800 - Accuracy: 0.755 | LearningRate: 0.00876 | StepTime: 49.827295s - Rate: 231,653.7 Tokens/s

2022-12-31 20:31:38

[ 18937/262144] Loss: 0.811 - Accuracy: 0.752 | LearningRate: 0.00876 | StepTime: 49.817297s - Rate: 231,568.0 Tokens/s

2022-12-31 20:32:29

[ 18941/262144] Loss: 0.806 - Accuracy: 0.754 | LearningRate: 0.00876 | StepTime: 49.832610s - Rate: 231,482.4 Tokens/s

2022-12-31 20:33:17

[ 18945/262144] Loss: 0.800 - Accuracy: 0.755 | LearningRate: 0.00876 | StepTime: 49.823160s - Rate: 231,396.8 Tokens/s

2022-12-31 20:34:07

[ 18949/262144] Loss: 0.808 - Accuracy: 0.752 | LearningRate: 0.00876 | StepTime: 49.823859s - Rate: 231,311.4 Tokens/s

2022-12-31 20:34:57

[ 18953/262144] Loss: 0.796 - Accuracy: 0.756 | LearningRate: 0.00876 | StepTime: 49.827651s - Rate: 231,226.1 Tokens/s

2022-12-31 20:35:47

[ 18957/262144] Loss: 0.813 - Accuracy: 0.751 | LearningRate: 0.00876 | StepTime: 49.814439s - Rate: 231,140.9 Tokens/s