Positive class (train): 495 examples
Positive class (test): 247 examples
Negative class (train): 495 examples
Negative class (test): 247 examples
Loading model 'Qwen/QwQ-32B' ...
Processing examples in batches...
Batch inference completed. Processed hidden states for positive examples for training subset.
Processing examples in batches...
Batch inference completed. Processed hidden states for positive examples for test subset.
Processing examples in batches...
Batch inference completed. Processed hidden states for negative examples for training subset.
Processing examples in batches...
Batch inference completed. Processed hidden states for negative examples for test subset.
Processing layer 0...
Epoch [100/100], Loss: 0.7210
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_0.pth
Accuracy: 0.6043715846994535
Processing layer 1...
Epoch [100/100], Loss: 0.1893
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_1.pth
Accuracy: 0.921311475409836
Processing layer 2...
Epoch [100/100], Loss: 0.1816
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_2.pth
Accuracy: 0.9180327868852459
Processing layer 3...
Epoch [100/100], Loss: 0.1698
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_3.pth
Accuracy: 0.9256830601092896
Processing layer 4...
Epoch [100/100], Loss: 0.1376
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_4.pth
Accuracy: 0.9333333333333333
Processing layer 5...
Epoch [100/100], Loss: 0.1114
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_5.pth
Accuracy: 0.9322404371584699
Processing layer 6...
Epoch [100/100], Loss: 0.0909
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_6.pth
Accuracy: 0.9420765027322404
Processing layer 7...
Epoch [100/100], Loss: 0.0686
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_7.pth
Accuracy: 0.940983606557377
Processing layer 8...
Epoch [100/100], Loss: 0.0619
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_8.pth
Accuracy: 0.9431693989071038
Processing layer 9...
Epoch [100/100], Loss: 0.0548
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_9.pth
Accuracy: 0.940983606557377
Processing layer 10...
Epoch [100/100], Loss: 0.0491
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_10.pth
Accuracy: 0.9398907103825137
Processing layer 11...
Epoch [100/100], Loss: 0.1862
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_11.pth
Accuracy: 0.9245901639344263
Processing layer 12...
Epoch [100/100], Loss: 0.1924
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_12.pth
Accuracy: 0.9289617486338798
Processing layer 13...
Epoch [100/100], Loss: 0.0815
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_13.pth
Accuracy: 0.9344262295081968
Processing layer 14...
Epoch [100/100], Loss: 0.0817
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_14.pth
Accuracy: 0.9333333333333333
Processing layer 15...
Epoch [100/100], Loss: 0.0471
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_15.pth
Accuracy: 0.9377049180327869
Processing layer 16...
Epoch [100/100], Loss: 0.0318
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_16.pth
Accuracy: 0.9442622950819672
Processing layer 17...
Epoch [100/100], Loss: 0.0116
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_17.pth
Accuracy: 0.9366120218579235
Processing layer 18...
Epoch [100/100], Loss: 0.0110
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_18.pth
Accuracy: 0.9387978142076503
Processing layer 19...
Epoch [100/100], Loss: 0.0112
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_19.pth
Accuracy: 0.9366120218579235
Processing layer 20...
Epoch [100/100], Loss: 0.0109
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_20.pth
Accuracy: 0.940983606557377
Processing layer 21...
Epoch [100/100], Loss: 0.0109
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_21.pth
Accuracy: 0.940983606557377
Processing layer 22...
Epoch [100/100], Loss: 0.0081
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_22.pth
Accuracy: 0.9387978142076503
Processing layer 23...
Epoch [100/100], Loss: 0.0060
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_23.pth
Accuracy: 0.9377049180327869
Processing layer 24...
Epoch [100/100], Loss: 0.0047
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_24.pth
Accuracy: 0.940983606557377
Processing layer 25...
Epoch [100/100], Loss: 0.0043
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_25.pth
Accuracy: 0.9355191256830601
Processing layer 26...
Epoch [100/100], Loss: 0.0031
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_26.pth
Accuracy: 0.9387978142076503
Processing layer 27...
Epoch [100/100], Loss: 0.0019
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_27.pth
Accuracy: 0.9442622950819672
Processing layer 28...
Epoch [100/100], Loss: 0.0019
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_28.pth
Accuracy: 0.9387978142076503
Processing layer 29...
Epoch [100/100], Loss: 0.0015
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_29.pth
Accuracy: 0.9420765027322404
Processing layer 30...
Epoch [100/100], Loss: 0.0013
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_30.pth
Accuracy: 0.940983606557377
Processing layer 31...
Epoch [100/100], Loss: 0.0010
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_31.pth
Accuracy: 0.9431693989071038
Processing layer 32...
Epoch [100/100], Loss: 0.0006
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_32.pth
Accuracy: 0.9431693989071038
Processing layer 33...
Epoch [100/100], Loss: 0.0019
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_33.pth
Accuracy: 0.940983606557377
Processing layer 34...
Epoch [100/100], Loss: 0.0034
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_34.pth
Accuracy: 0.9377049180327869
Processing layer 35...
Epoch [100/100], Loss: 0.0019
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_35.pth
Accuracy: 0.9387978142076503
Processing layer 36...
Epoch [100/100], Loss: 0.0014
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_36.pth
Accuracy: 0.9387978142076503
Processing layer 37...
Epoch [100/100], Loss: 0.0011
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_37.pth
Accuracy: 0.9442622950819672
Processing layer 38...
Epoch [100/100], Loss: 0.0006
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_38.pth
Accuracy: 0.9387978142076503
Processing layer 39...
Epoch [100/100], Loss: 0.0001
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_39.pth
Accuracy: 0.9442622950819672
Processing layer 40...
Epoch [100/100], Loss: 0.0009
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_40.pth
Accuracy: 0.946448087431694
Processing layer 41...
Epoch [100/100], Loss: 0.0013
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_41.pth
Accuracy: 0.9486338797814208
Processing layer 42...
Epoch [100/100], Loss: 0.0016
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_42.pth
Accuracy: 0.946448087431694
Processing layer 43...
Epoch [100/100], Loss: 0.0012
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_43.pth
Accuracy: 0.940983606557377
Processing layer 44...
Epoch [100/100], Loss: 0.0000
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_44.pth
Accuracy: 0.946448087431694
Processing layer 45...
Epoch [100/100], Loss: 0.0000
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_45.pth
Accuracy: 0.9442622950819672
Processing layer 46...
Epoch [100/100], Loss: 0.0000
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_46.pth
Accuracy: 0.9497267759562842
Processing layer 47...
Epoch [100/100], Loss: 0.0000
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_47.pth
Accuracy: 0.9519125683060109
Processing layer 48...
Epoch [100/100], Loss: 0.0000
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_48.pth
Accuracy: 0.9475409836065574
Processing layer 49...
Epoch [100/100], Loss: 0.0000
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_49.pth
Accuracy: 0.9497267759562842
Processing layer 50...
Epoch [100/100], Loss: 0.0000
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_50.pth
Accuracy: 0.9486338797814208
Processing layer 51...
Epoch [100/100], Loss: 0.0104
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_51.pth
Accuracy: 0.9453551912568307
Processing layer 52...
Epoch [100/100], Loss: 0.0000
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_52.pth
Accuracy: 0.9431693989071038
Processing layer 53...
Epoch [100/100], Loss: 0.1346
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_53.pth
Accuracy: 0.9431693989071038
Processing layer 54...
Epoch [100/100], Loss: 0.2718
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_54.pth
Accuracy: 0.9453551912568307
Processing layer 55...
Epoch [100/100], Loss: 1.0442
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_55.pth
Accuracy: 0.9300546448087431
Processing layer 56...
Epoch [100/100], Loss: 0.0003
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_56.pth
Accuracy: 0.946448087431694
Processing layer 57...
Epoch [100/100], Loss: 0.0000
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_57.pth
Accuracy: 0.9508196721311475
Processing layer 58...
Epoch [100/100], Loss: 0.0000
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_58.pth
Accuracy: 0.9486338797814208
Processing layer 59...
Epoch [100/100], Loss: 1.0824
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_59.pth
Accuracy: 0.9475409836065574
Processing layer 60...
Epoch [100/100], Loss: 1.6293
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_60.pth
Accuracy: 0.9420765027322404
Processing layer 61...
Epoch [100/100], Loss: 2.8860
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_61.pth
Accuracy: 0.9278688524590164
Processing layer 62...
Epoch [100/100], Loss: 0.0000
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_62.pth
Accuracy: 0.9540983606557377
Processing layer 63...
Epoch [100/100], Loss: 7.2391
Training complete.
Model saved to from_evidence_negative_awareness_positive_awareness_avg_mlp/model_63.pth
Accuracy: 0.940983606557377
