README.md

metadata

library_name: transformers
license: apache-2.0
base_model: allenai/led-base-16384
tags:
  - generated_from_trainer
metrics:
  - rouge
  - bleu
  - precision
  - recall
  - f1
model-index:
  - name: LED_sum_challenge2
    results: []

LED_sum_challenge2

This model is a fine-tuned version of allenai/led-base-16384 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.9586
• Rouge1: 0.2918
• Rouge2: 0.1012
• Rougel: 0.2293
• Rougelsum: 0.2288
• Gen Len: 28.12
• Bleu: 0.0548
• Precisions: 0.1048
• Brevity Penalty: 0.9001
• Length Ratio: 0.9048
• Translation Length: 1093.0
• Reference Length: 1208.0
• Precision: 0.8818
• Recall: 0.8759
• F1: 0.8788
• Hashcode: roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• 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	Bleu	Precisions	Brevity Penalty	Length Ratio	Translation Length	Reference Length	Precision	Recall	F1	Hashcode
9.0848	1.0	13	7.5283	0.24	0.0579	0.1713	0.1714	31.78	0.0296	0.0629	1.0	1.0439	1261.0	1208.0	0.8521	0.8597	0.8558	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
6.171	2.0	26	4.9217	0.2695	0.0854	0.203	0.2033	25.98	0.0368	0.0987	0.8063	0.8228	994.0	1208.0	0.8806	0.8705	0.8755	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
4.4536	3.0	39	4.1312	0.2717	0.0862	0.2162	0.2157	23.34	0.0352	0.1067	0.6694	0.7136	862.0	1208.0	0.8846	0.8732	0.8788	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
3.7683	4.0	52	3.7332	0.3043	0.0981	0.2301	0.2308	25.46	0.0499	0.1154	0.7784	0.7997	966.0	1208.0	0.8885	0.8787	0.8835	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
3.3278	5.0	65	3.4699	0.2978	0.1041	0.2351	0.2344	25.38	0.0497	0.1117	0.7854	0.8055	973.0	1208.0	0.8869	0.8763	0.8815	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
3.0332	6.0	78	3.2808	0.2946	0.1013	0.2335	0.2319	26.48	0.0503	0.1069	0.8181	0.8328	1006.0	1208.0	0.8857	0.8774	0.8815	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
2.8037	7.0	91	3.1443	0.295	0.0965	0.2275	0.2264	27.52	0.0428	0.0978	0.8612	0.87	1051.0	1208.0	0.8822	0.8777	0.8799	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
2.637	8.0	104	3.0523	0.2834	0.0997	0.2263	0.2257	27.22	0.0499	0.1034	0.8527	0.8626	1042.0	1208.0	0.8813	0.8752	0.8781	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
2.5158	9.0	117	2.9900	0.2821	0.0989	0.2271	0.2273	27.18	0.0508	0.1051	0.848	0.8584	1037.0	1208.0	0.8842	0.8773	0.8806	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
2.4321	10.0	130	2.9586	0.2918	0.1012	0.2293	0.2288	28.12	0.0548	0.1048	0.9001	0.9048	1093.0	1208.0	0.8818	0.8759	0.8788	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)

Framework versions

• Transformers 4.53.1
• Pytorch 2.7.0+cu126
• Datasets 3.6.0
• Tokenizers 0.21.1