0ff38d7ffdb6961f868e058d5b3106c8

This model is a fine-tuned version of facebook/opt-125m on the ccdv/patent-classification [abstract] dataset. It achieves the following results on the evaluation set:

• Loss: 1.7756
• Data Size: 1.0
• Epoch Runtime: 59.2126
• Accuracy: 0.6256
• F1 Macro: 0.5560

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 32
• total_eval_batch_size: 32
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Accuracy	F1 Macro
No log	0	0	2.3848	0	4.6216	0.1148	0.0497
No log	1	781	1.8682	0.0078	5.1967	0.2897	0.1436
No log	2	1562	1.5922	0.0156	5.4509	0.3672	0.2259
No log	3	2343	1.3406	0.0312	6.3451	0.5152	0.3779
0.0351	4	3124	1.2140	0.0625	7.9920	0.5723	0.4668
1.1674	5	3905	1.1442	0.125	11.5759	0.5855	0.5189
1.1097	6	4686	1.0385	0.25	19.0785	0.6348	0.5680
0.9667	7	5467	1.0301	0.5	31.8041	0.6462	0.5856
0.8739	8.0	6248	0.9999	1.0	59.0072	0.6506	0.5970
0.6974	9.0	7029	1.0473	1.0	61.3541	0.6518	0.5853
0.4311	10.0	7810	1.3465	1.0	59.1640	0.6328	0.5947
0.2973	11.0	8591	1.5955	1.0	58.9765	0.6288	0.5779
0.2243	12.0	9372	1.7756	1.0	59.2126	0.6256	0.5560

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.3.0
• Tokenizers 0.22.1