LLM_Teached_PEGASUS_CNNDM_2

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.7016
• Rouge1: 0.4651
• Rouge2: 0.2076
• Rougel: 0.3457
• Rougelsum: 0.3459
• Gen Len: 52.1582
• Precision: 0.906
• Recall: 0.9098
• F1: 0.9077

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len	Precision	Recall	F1
No log	1.0	312	1.7705	0.4551	0.1985	0.335	0.3351	51.6464	0.9043	0.9073	0.9056
1.8539	2.0	625	1.7468	0.4578	0.2016	0.3394	0.3397	51.0627	0.9054	0.908	0.9065
1.8539	3.0	937	1.7331	0.4595	0.2019	0.3389	0.3391	52.9318	0.9039	0.9089	0.9063
1.7903	4.0	1250	1.7226	0.4606	0.2032	0.3406	0.3405	52.8055	0.9046	0.9094	0.9068
1.746	5.0	1562	1.7132	0.4642	0.2068	0.3453	0.3453	51.7873	0.9062	0.9096	0.9077
1.746	6.0	1875	1.7117	0.463	0.2055	0.3435	0.3436	53.4382	0.905	0.91	0.9073
1.7173	7.0	2187	1.7057	0.4644	0.2073	0.3456	0.3457	52.1718	0.906	0.9099	0.9078
1.7004	8.0	2500	1.7033	0.4668	0.2084	0.3464	0.3466	51.9	0.9063	0.91	0.908
1.7004	9.0	2812	1.7027	0.4651	0.2074	0.3457	0.3458	52.3591	0.906	0.9099	0.9078
1.6888	9.98	3120	1.7016	0.4651	0.2076	0.3457	0.3459	52.1582	0.906	0.9098	0.9077

Framework versions

• Transformers 4.36.0
• Pytorch 2.0.1+cu117
• Datasets 2.7.1
• Tokenizers 0.15.2