edaf807035033ae550de85b6e9f46fc8

This model is a fine-tuned version of meta-llama/Llama-3.2-1B on the nyu-mll/glue [cola] dataset. It achieves the following results on the evaluation set:

• Loss: 2.8355
• Data Size: 1.0
• Epoch Runtime: 39.3285
• Accuracy: 0.6816
• F1 Macro: 0.4955
• Rouge1: 0.6826
• Rouge2: 0.0
• Rougel: 0.6816
• Rougelsum: 0.6816

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	Rouge1	Rouge2	Rougel	Rougelsum
No log	0	0	3.7470	0	2.5636	0.4990	0.4971	0.4980	0.0	0.5	0.4990
No log	1	267	17.4966	0.0078	2.7785	0.6875	0.4074	0.6885	0.0	0.6885	0.6875
No log	2	534	2.5784	0.0156	3.9404	0.6436	0.5012	0.6426	0.0	0.6445	0.6436
No log	3	801	5.6768	0.0312	6.1631	0.6406	0.5008	0.6406	0.0	0.6396	0.6406
No log	4	1068	2.5512	0.0625	7.4907	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
0.3145	5	1335	2.8096	0.125	10.2584	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
3.323	6	1602	4.2677	0.25	14.7420	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
2.8374	7	1869	2.4697	0.5	23.8492	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
2.5821	8.0	2136	2.5515	1.0	42.5746	0.6602	0.5224	0.6611	0.0	0.6602	0.6611
2.3962	9.0	2403	2.5414	1.0	41.4768	0.6592	0.5596	0.6592	0.0	0.6592	0.6592
2.1462	10.0	2670	3.1376	1.0	41.7784	0.6855	0.5178	0.6855	0.0	0.6855	0.6855
1.9416	11.0	2937	2.8355	1.0	39.3285	0.6816	0.4955	0.6826	0.0	0.6816	0.6816

Framework versions

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