13508a910088b10648dcbd4446beefbb

This model is a fine-tuned version of deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B on the nyu-mll/glue dataset. It achieves the following results on the evaluation set:

• Loss: 2.0217
• Data Size: 1.0
• Epoch Runtime: 444.8703
• Accuracy: 0.8773
• F1 Macro: 0.8772
• Rouge1: 0.8773
• Rouge2: 0.0
• Rougel: 0.8773
• Rougelsum: 0.8773

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	5.0392	0	2.7975	0.4757	0.3508	0.4769	0.0	0.4769	0.4757
No log	1	2104	3.6289	0.0078	6.4593	0.6771	0.6401	0.6771	0.0	0.6771	0.6771
No log	2	4208	1.9409	0.0156	10.2146	0.7928	0.7919	0.7928	0.0	0.7928	0.7928
0.0385	3	6312	1.1442	0.0312	19.1107	0.8866	0.8865	0.8866	0.0	0.8866	0.8866
1.2209	4	8416	1.1750	0.0625	33.0889	0.8785	0.8781	0.8785	0.0	0.8785	0.8785
0.9629	5	10520	1.0499	0.125	62.0677	0.8912	0.8909	0.8900	0.0	0.8912	0.8912
0.7813	6	12624	1.1324	0.25	115.9244	0.8796	0.8793	0.8796	0.0	0.8796	0.8796
0.7602	7	14728	1.0301	0.5	224.1109	0.8981	0.8981	0.8981	0.0	0.8981	0.8981
0.555	8.0	16832	1.1674	1.0	444.0133	0.8900	0.8900	0.8900	0.0	0.8900	0.8900
0.4034	9.0	18936	1.6610	1.0	440.9769	0.8900	0.8900	0.8900	0.0	0.8900	0.8900
0.3886	10.0	21040	1.8956	1.0	446.9196	0.8785	0.8782	0.8785	0.0	0.8785	0.8785
0.2786	11.0	23144	2.0217	1.0	444.8703	0.8773	0.8772	0.8773	0.0	0.8773	0.8773

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.1