metadata
license: apache-2.0
base_model: allenai/led-base-16384
tags:
- generated_from_trainer
datasets:
- big_patent
metrics:
- rouge
model-index:
- name: led-base-big-patent
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: big_patent
type: big_patent
config: g
split: validation
args: g
metrics:
- name: Rouge1
type: rouge
value: 0.2658
led-base-big-patent
This model is a fine-tuned version of allenai/led-base-16384 on the big_patent dataset. It achieves the following results on the evaluation set:
- Loss: 0.2665
- Rouge1: 0.2658
- Rouge2: 0.1008
- Rougel: 0.2231
- Rougelsum: 0.2244
- Gen Len: 19.6593
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: 4.892476e-05
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 108 | 0.2940 | 0.2525 | 0.0809 | 0.2062 | 0.2074 | 19.956 |
| No log | 2.0 | 216 | 0.2595 | 0.2713 | 0.097 | 0.2233 | 0.2256 | 19.5165 |
| No log | 3.0 | 324 | 0.2665 | 0.2658 | 0.1008 | 0.2231 | 0.2244 | 19.6593 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2