owanr/r1_iterater

output_r1_iter_wo_p

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

• Loss: 0.1334
• Bleu: 0.0
• Gen Len: 2.432

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 4
• total_train_batch_size: 512
• total_eval_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20.0

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
No log	1.0	27	0.2728	0.0	2.9953
No log	2.0	54	0.2650	0.0	2.6791
No log	3.0	81	0.2637	0.0	2.1874
No log	4.0	108	0.2418	0.0	2.2973
No log	5.0	135	0.2738	0.0	2.2494
No log	6.0	162	0.1914	0.0	2.3812
No log	7.0	189	0.1641	0.0	2.3983
No log	8.0	216	0.1695	0.0	2.3995
No log	9.0	243	0.1521	0.0	2.4167
No log	10.0	270	0.1569	0.0	2.4167
No log	11.0	297	0.1615	0.0	2.4137
No log	12.0	324	0.1473	0.0	2.4238
No log	13.0	351	0.1376	0.0	2.4255
No log	14.0	378	0.1495	0.0	2.419
No log	15.0	405	0.1334	0.0	2.432
No log	16.0	432	0.1474	0.0	2.4214
No log	17.0	459	0.1484	0.0	2.4291
No log	18.0	486	0.1407	0.0	2.4297
0.1905	19.0	513	0.1568	0.0	2.4208
0.1905	20.0	540	0.1631	0.0	2.4261

Framework versions

• Transformers 4.30.2
• Pytorch 1.13.1+cu117
• Datasets 2.12.0
• Tokenizers 0.13.3