LLM_Teached_Bart_50k

This model is a fine-tuned version of facebook/bart-large-xsum on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.5590
• Rouge1: 0.4909
• Rouge2: 0.2303
• Rougel: 0.3967
• Rougelsum: 0.3965
• Gen Len: 38.2287
• Precision: 0.9063
• Recall: 0.9187
• F1: 0.9123

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: 32
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4
• 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	390	1.6214	0.4804	0.2218	0.3873	0.3873	38.3549	0.9049	0.9166	0.9106
1.5842	2.0	781	1.5548	0.4874	0.2283	0.3945	0.3945	37.8604	0.9059	0.9171	0.9113
1.3014	3.0	1172	1.5461	0.49	0.2294	0.3975	0.3974	37.7564	0.9064	0.918	0.912
1.18	3.99	1560	1.5590	0.4909	0.2303	0.3967	0.3965	38.2287	0.9063	0.9187	0.9123

Framework versions

• Transformers 4.36.0
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.15.0