homebrewnlp

Clashluke's group workspace

Group: rnn-shortctx-nosign

3560

300

308

446

1-4

of 4

Timestamps visible

2023-05-10 14:17:49

[ 3272/65536] Loss: 3.460 - Accuracy: 0.166 | LearningRate: 0.00098 | StepTime: 5.136694s - Rate: 5,417.8 Tokens/s

2023-05-10 14:17:55

[ 3273/65536] Loss: 3.429 - Accuracy: 0.172 | LearningRate: 0.00098 | StepTime: 5.137995s - Rate: 5,418.1 Tokens/s

2023-05-10 14:17:59

[ 3274/65536] Loss: 3.433 - Accuracy: 0.169 | LearningRate: 0.00098 | StepTime: 5.133432s - Rate: 5,418.3 Tokens/s

2023-05-10 14:18:05

[ 3275/65536] Loss: 3.416 - Accuracy: 0.168 | LearningRate: 0.00098 | StepTime: 5.133345s - Rate: 5,418.5 Tokens/s

2023-05-10 14:18:09

[ 3276/65536] Loss: 3.438 - Accuracy: 0.164 | LearningRate: 0.00098 | StepTime: 5.139072s - Rate: 5,418.8 Tokens/s

2023-05-10 14:18:15

[ 3277/65536] Loss: 3.412 - Accuracy: 0.169 | LearningRate: 0.00098 | StepTime: 5.135076s - Rate: 5,419.0 Tokens/s

2023-05-10 14:18:19

[ 3278/65536] Loss: 3.466 - Accuracy: 0.163 | LearningRate: 0.00098 | StepTime: 5.136409s - Rate: 5,419.3 Tokens/s

2023-05-10 14:18:25

[ 3279/65536] Loss: 3.412 - Accuracy: 0.171 | LearningRate: 0.00098 | StepTime: 5.136999s - Rate: 5,419.5 Tokens/s

2023-05-10 14:18:31

[ 3280/65536] Loss: 3.461 - Accuracy: 0.170 | LearningRate: 0.00098 | StepTime: 5.141619s - Rate: 5,419.8 Tokens/s

2023-05-10 14:18:35

[ 3281/65536] Loss: 3.443 - Accuracy: 0.161 | LearningRate: 0.00098 | StepTime: 5.136311s - Rate: 5,420.0 Tokens/s

2023-05-10 14:18:41

[ 3282/65536] Loss: 3.576 - Accuracy: 0.162 | LearningRate: 0.00098 | StepTime: 5.138113s - Rate: 5,420.3 Tokens/s

2023-05-10 14:18:45

[ 3283/65536] Loss: 3.487 - Accuracy: 0.163 | LearningRate: 0.00098 | StepTime: 5.133212s - Rate: 5,420.5 Tokens/s