metadata
library_name: peft
license: apache-2.0
base_model: allenai/led-base-16384
tags:
- generated_from_trainer
metrics:
- rouge
- bleu
- precision
- recall
- f1
model-index:
- name: Lora_LED_sum_challenge
results: []
Lora_LED_sum_challenge
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: 4.1010
- Rouge1: 0.3015
- Rouge2: 0.0982
- Rougel: 0.2325
- Rougelsum: 0.234
- Gen Len: 27.86
- Bleu: 0.0493
- Precisions: 0.1077
- Brevity Penalty: 0.8669
- Length Ratio: 0.875
- Translation Length: 1057.0
- Reference Length: 1208.0
- Precision: 0.8803
- Recall: 0.8763
- F1: 0.8783
- 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: 0.002
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 16
- 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
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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 8.1516 | 1.0 | 7 | 7.5736 | 0.2121 | 0.0491 | 0.1581 | 0.1587 | 32.0 | 0.0174 | 0.0574 | 1.0 | 1.0728 | 1296.0 | 1208.0 | 0.8534 | 0.8583 | 0.8557 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) |
| 5.8141 | 2.0 | 14 | 5.0888 | 0.2526 | 0.0765 | 0.1991 | 0.2004 | 26.88 | 0.0316 | 0.0882 | 0.822 | 0.8361 | 1010.0 | 1208.0 | 0.8772 | 0.8715 | 0.8742 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) |
| 4.3777 | 3.0 | 21 | 4.4191 | 0.2668 | 0.0907 | 0.2057 | 0.2072 | 24.04 | 0.0421 | 0.1088 | 0.7134 | 0.7475 | 903.0 | 1208.0 | 0.8824 | 0.8719 | 0.877 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) |
| 3.9067 | 4.0 | 28 | 4.2179 | 0.2684 | 0.0813 | 0.2084 | 0.2085 | 25.14 | 0.0378 | 0.1006 | 0.7488 | 0.7757 | 937.0 | 1208.0 | 0.8799 | 0.8705 | 0.8751 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) |
| 3.6847 | 5.0 | 35 | 4.1231 | 0.2897 | 0.0861 | 0.2227 | 0.2226 | 29.34 | 0.0362 | 0.0876 | 0.9412 | 0.9429 | 1139.0 | 1208.0 | 0.8751 | 0.8761 | 0.8756 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) |
| 3.5317 | 6.0 | 42 | 4.1113 | 0.2644 | 0.0858 | 0.2097 | 0.2107 | 26.66 | 0.0395 | 0.0983 | 0.8063 | 0.8228 | 994.0 | 1208.0 | 0.8826 | 0.8744 | 0.8784 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) |
| 3.4303 | 7.0 | 49 | 4.0934 | 0.2866 | 0.0945 | 0.2219 | 0.2226 | 27.02 | 0.0407 | 0.1017 | 0.8413 | 0.8526 | 1030.0 | 1208.0 | 0.8827 | 0.8773 | 0.8799 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) |
| 3.3587 | 8.0 | 56 | 4.0800 | 0.2956 | 0.1007 | 0.2287 | 0.2302 | 28.1 | 0.0467 | 0.1031 | 0.8734 | 0.8808 | 1064.0 | 1208.0 | 0.8805 | 0.8756 | 0.878 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) |
| 3.3033 | 9.0 | 63 | 4.0926 | 0.2924 | 0.0982 | 0.2205 | 0.2225 | 27.06 | 0.0481 | 0.1062 | 0.8461 | 0.8568 | 1035.0 | 1208.0 | 0.8813 | 0.8747 | 0.8779 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) |
| 3.2828 | 10.0 | 70 | 4.1010 | 0.3015 | 0.0982 | 0.2325 | 0.234 | 27.86 | 0.0493 | 0.1077 | 0.8669 | 0.875 | 1057.0 | 1208.0 | 0.8803 | 0.8763 | 0.8783 | roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1) |
Framework versions
- PEFT 0.15.2
- Transformers 4.53.1
- Pytorch 2.7.0+cu126
- Datasets 3.6.0
- Tokenizers 0.21.1