t5-small-finetuned-manimml-1.1

This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0750
• Rouge1: 26.2677
• Rouge2: 21.24
• Rougel: 26.2649
• Rougelsum: 26.2291
• Gen Len: 19.0

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	50	1.6688	4.48	0.3941	4.3103	4.3263	19.0
No log	2.0	100	0.8312	11.3557	2.7306	10.6554	10.6787	18.985
No log	3.0	150	0.4371	24.1056	14.6464	23.4038	23.3396	19.0
No log	4.0	200	0.2658	26.1867	17.1716	25.6324	25.5875	19.0
No log	5.0	250	0.1862	25.7525	18.2023	25.3068	25.2656	19.0
No log	6.0	300	0.1429	26.1909	19.8811	26.0053	26.0106	19.0
No log	7.0	350	0.1180	26.1977	20.6534	26.1067	26.0828	19.0
No log	8.0	400	0.1055	26.0723	21.0154	26.0943	26.0582	19.0
No log	9.0	450	0.0934	26.2485	21.1998	26.2506	26.2173	19.0
0.8503	10.0	500	0.0863	26.2677	21.24	26.2649	26.2291	19.0
0.8503	11.0	550	0.0819	26.2677	21.24	26.2649	26.2291	19.0
0.8503	12.0	600	0.0795	26.2677	21.24	26.2649	26.2291	19.0
0.8503	13.0	650	0.0768	26.2677	21.24	26.2649	26.2291	19.0
0.8503	14.0	700	0.0755	26.2677	21.24	26.2649	26.2291	19.0
0.8503	15.0	750	0.0750	26.2677	21.24	26.2649	26.2291	19.0

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1