b81db5edc1723338fe5e69ea9599c43f

This model is a fine-tuned version of facebook/opt-125m on the nyu-mll/glue [cola] dataset. It achieves the following results on the evaluation set:

• Loss: 0.8617
• Data Size: 1.0
• Epoch Runtime: 20.0130
• Accuracy: 0.7607
• F1 Macro: 0.7223
• Rouge1: 0.7607
• Rouge2: 0.0
• Rougel: 0.7607
• Rougelsum: 0.7617

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	0.7581	0	1.6190	0.6885	0.4137	0.6890	0.0	0.6885	0.6885
No log	1	267	0.6682	0.0078	2.4625	0.6201	0.5079	0.6211	0.0	0.6211	0.6201
No log	2	534	0.7250	0.0156	1.8983	0.6885	0.4137	0.6895	0.0	0.6895	0.6885
No log	3	801	0.6312	0.0312	2.2904	0.6689	0.5196	0.6699	0.0	0.6689	0.6689
No log	4	1068	0.6133	0.0625	3.0742	0.6963	0.4605	0.6963	0.0	0.6963	0.6963
0.0368	5	1335	0.7332	0.125	4.2057	0.6895	0.4111	0.6904	0.0	0.6899	0.6895
0.5281	6	1602	0.5788	0.25	6.3516	0.7236	0.5645	0.7246	0.0	0.7227	0.7236
0.472	7	1869	0.5995	0.5	10.9350	0.75	0.6258	0.75	0.0	0.75	0.75
0.3882	8.0	2136	0.5349	1.0	20.0279	0.7578	0.6946	0.7578	0.0	0.7588	0.7578
0.2541	9.0	2403	0.5746	1.0	19.8151	0.7598	0.7041	0.7598	0.0	0.7598	0.7607
0.1935	10.0	2670	0.8062	1.0	19.7531	0.7568	0.6930	0.7568	0.0	0.7568	0.7578
0.1755	11.0	2937	0.7620	1.0	20.0056	0.7529	0.7015	0.7520	0.0	0.7529	0.7529
0.1829	12.0	3204	0.8617	1.0	20.0130	0.7607	0.7223	0.7607	0.0	0.7607	0.7617

Framework versions

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