peft-InstructionTuning-training-1719334832

This model is a fine-tuned version of google/flan-t5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: nan
• Rouge1: 6.2665
• Rouge2: 0.0511
• Rougel: 5.2669
• Rougelsum: 5.4181

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: 6
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 12
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 50
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
0.0	2.6316	50	nan	6.2665	0.0511	5.2669	5.4181
0.0	5.2632	100	nan	6.2665	0.0511	5.2669	5.4181
0.0	7.8947	150	nan	6.2665	0.0511	5.2669	5.4181
0.0	10.5263	200	nan	6.2665	0.0511	5.2669	5.4181
0.0	13.1579	250	nan	6.2665	0.0511	5.2669	5.4181
0.0	15.7895	300	nan	6.2665	0.0511	5.2669	5.4181
0.0	18.4211	350	nan	6.2665	0.0511	5.2669	5.4181
0.0	21.0526	400	nan	6.2665	0.0511	5.2669	5.4181
0.0	23.6842	450	nan	6.2665	0.0511	5.2669	5.4181
0.0	26.3158	500	nan	6.2665	0.0511	5.2669	5.4181
0.0	28.9474	550	nan	6.2665	0.0511	5.2669	5.4181
0.0	31.5789	600	nan	6.2665	0.0511	5.2669	5.4181
0.0	34.2105	650	nan	6.2665	0.0511	5.2669	5.4181
0.0	36.8421	700	nan	6.2665	0.0511	5.2669	5.4181
0.0	39.4737	750	nan	6.2665	0.0511	5.2669	5.4181
0.0	42.1053	800	nan	6.2665	0.0511	5.2669	5.4181
0.0	44.7368	850	nan	6.2665	0.0511	5.2669	5.4181
0.0	47.3684	900	nan	6.2665	0.0511	5.2669	5.4181
0.0	50.0	950	nan	6.2665	0.0511	5.2669	5.4181

Framework versions

• PEFT 0.11.1
• Transformers 4.41.2
• Pytorch 2.1.2
• Datasets 2.19.2
• Tokenizers 0.19.1