learning_rate: 3e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss
13.5664	0.5714	1	14.0489
18.6422	1.5714	2	13.0300
17.8287	2.5714	3	12.0308
17.0632	3.5714	4	11.0754
16.1516	4.5714	5	10.1440
14.6881	5.5714	6	9.1360
14.226	6.5714	7	8.1785
12.7129	7.5714	8	7.4610
11.9365	8.5714	9	6.7690
10.6449	9.5714	10	6.0893
9.8716	10.5714	11	5.4196
9.1578	11.5714	12	4.7783
9.1422	12.5714	13	4.1559
8.0235	13.5714	14	3.5585
7.2521	14.5714	15	3.0355
6.9468	15.5714	16	2.5595
6.3573	16.5714	17	2.1188
6.0266	17.5714	18	1.7322
5.9024	18.5714	19	1.3607
4.5874	19.5714	20	1.0799
4.7103	20.5714	21	0.9026
4.5861	21.5714	22	0.8025
4.1161	22.5714	23	0.7488
3.805	23.5714	24	0.7257
3.3272	24.5714	25	0.7164
3.2706	25.5714	26	0.7117
3.368	26.5714	27	0.7095
3.3499	27.5714	28	0.7062
2.9882	28.5714	29	0.7011
2.798	29.5714	30	0.6939
2.7807	30.5714	31	0.6856
2.6064	31.5714	32	0.6774
2.537	32.5714	33	0.6680
2.5005	33.5714	34	0.6602
2.4666	34.5714	35	0.6530
2.919	35.5714	36	0.6471
2.3748	36.5714	37	0.6416
2.3416	37.5714	38	0.6363
2.3233	38.5714	39	0.6312
2.2107	39.5714	40	0.6265
2.2191	40.5714	41	0.6220
2.1931	41.5714	42	0.6178
2.0958	42.5714	43	0.6136
1.9957	43.5714	44	0.6095
2.0614	44.5714	45	0.6055
2.0534	45.5714	46	0.6019
2.0221	46.5714	47	0.5987
1.8483	47.5714	48	0.5957
1.9068	48.5714	49	0.5929
1.9266	49.5714	50	0.5903
1.8266	50.5714	51	0.5882
1.7679	51.5714	52	0.5864
1.7766	52.5714	53	0.5851
1.7592	53.5714	54	0.5840
1.8102	54.5714	55	0.5832
1.6658	55.5714	56	0.5826
1.6261	56.5714	57	0.5820
1.6287	57.5714	58	0.5816
1.6453	58.5714	59	0.5814
1.5957	59.5714	60	0.5813
1.5529	60.5714	61	0.5813
1.524	61.5714	62	0.5814
1.5488	62.5714	63	0.5815
1.4817	63.5714	64	0.5817
1.5539	64.5714	65	0.5820
1.4465	65.5714	66	0.5821
1.4489	66.5714	67	0.5820
1.4475	67.5714	68	0.5820
1.4058	68.5714	69	0.5819
1.347	69.5714	70	0.5817
1.42	70.5714	71	0.5815
1.3531	71.5714	72	0.5813
1.5894	72.5714	73	0.5811
1.3858	73.5714	74	0.5808
1.3335	74.5714	75	0.5802
1.3272	75.5714	76	0.5796
1.3593	76.5714	77	0.5791
1.2942	77.5714	78	0.5785
1.3547	78.5714	79	0.5778
1.3148	79.5714	80	0.5767
1.3101	80.5714	81	0.5757
1.2415	81.5714	82	0.5747
1.2511	82.5714	83	0.5738
1.324	83.5714	84	0.5730
1.2649	84.5714	85	0.5723
1.2953	85.5714	86	0.5717
1.2885	86.5714	87	0.5711
1.2763	87.5714	88	0.5704
1.3113	88.5714	89	0.5698
1.3509	89.5714	90	0.5694
1.2008	90.5714	91	0.5689
1.2398	91.5714	92	0.5685
1.2502	92.5714	93	0.5680
1.2558	93.5714	94	0.5676
1.2674	94.5714	95	0.5672
1.1993	95.5714	96	0.5668
1.3029	96.5714	97	0.5665
1.2861	97.5714	98	0.5662
1.2414	98.5714	99	0.5661
1.2446	99.5714	100	0.5660

Framework versions

Transformers 4.48.3
Pytorch 2.5.1+cu124
Datasets 3.3.0
Tokenizers 0.21.0

devagonal/t5-small-squad-qag

t5-small-squad-qag

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

devagonal/
t5-small-squad-qag