flan-t5-base-gsm8k_model

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: 0.8620
• Rouge1: 0.2697
• Rouge2: 0.1356
• Rougel: 0.2253
• Rougelsum: 0.2422

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: 0.0005
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
1.1948	1.0	654	0.9260	0.2684	0.1280	0.2208	0.2394
0.9443	2.0	1308	0.8731	0.2686	0.1332	0.2242	0.2419
0.8297	3.0	1962	0.8620	0.2697	0.1356	0.2253	0.2422
0.6273	4.0	2616	0.8873	0.2732	0.1404	0.2289	0.2461
0.558	5.0	3270	0.8902	0.2736	0.1426	0.2300	0.2464
0.499	6.0	3924	0.9316	0.2750	0.1424	0.2305	0.2479
0.4007	7.0	4578	0.9573	0.2777	0.1437	0.2322	0.2489
0.353	8.0	5232	1.0082	0.2743	0.1431	0.2308	0.2478
0.3235	9.0	5886	1.0506	0.2761	0.1463	0.2332	0.2499
0.2644	10.0	6540	1.1053	0.2780	0.1465	0.2338	0.2509
0.2325	11.0	7194	1.1463	0.2785	0.1478	0.2345	0.2513
0.2171	12.0	7848	1.2101	0.2784	0.1492	0.2344	0.2518
0.1852	13.0	8502	1.2566	0.2784	0.1470	0.2336	0.2506
0.1692	14.0	9156	1.3192	0.2777	0.1463	0.2325	0.2498
0.1602	15.0	9810	1.3562	0.2787	0.1473	0.2333	0.2507

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1