t5-base-squad-qag

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

• Loss: 0.1945

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: 8
• eval_batch_size: 8
• seed: 42
• 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
No log	1.0	7	12.2471
No log	2.0	14	7.2702
No log	3.0	21	5.6811
No log	4.0	28	4.6100
No log	5.0	35	0.6711
No log	6.0	42	0.4312
No log	7.0	49	0.4167
No log	8.0	56	0.4011
No log	9.0	63	0.3785
No log	10.0	70	0.3256
No log	11.0	77	0.2868
No log	12.0	84	0.2607
No log	13.0	91	0.2423
No log	14.0	98	0.2277
No log	15.0	105	0.2053
No log	16.0	112	0.1962
No log	17.0	119	0.1866
No log	18.0	126	0.1822
No log	19.0	133	0.1796
No log	20.0	140	0.1789
No log	21.0	147	0.1782
No log	22.0	154	0.1774
No log	23.0	161	0.1760
No log	24.0	168	0.1754
No log	25.0	175	0.1754
No log	26.0	182	0.1748
No log	27.0	189	0.1739
No log	28.0	196	0.1730
No log	29.0	203	0.1728
No log	30.0	210	0.1728
No log	31.0	217	0.1734
No log	32.0	224	0.1736
No log	33.0	231	0.1733
No log	34.0	238	0.1731
No log	35.0	245	0.1738
No log	36.0	252	0.1744
No log	37.0	259	0.1747
No log	38.0	266	0.1745
No log	39.0	273	0.1739
No log	40.0	280	0.1747
No log	41.0	287	0.1752
No log	42.0	294	0.1757
No log	43.0	301	0.1768
No log	44.0	308	0.1776
No log	45.0	315	0.1787
No log	46.0	322	0.1800
No log	47.0	329	0.1799
No log	48.0	336	0.1801
No log	49.0	343	0.1801
No log	50.0	350	0.1808
No log	51.0	357	0.1827
No log	52.0	364	0.1842
No log	53.0	371	0.1839
No log	54.0	378	0.1841
No log	55.0	385	0.1844
No log	56.0	392	0.1835
No log	57.0	399	0.1835
No log	58.0	406	0.1839
No log	59.0	413	0.1837
No log	60.0	420	0.1838
No log	61.0	427	0.1841
No log	62.0	434	0.1846
No log	63.0	441	0.1849
No log	64.0	448	0.1857
No log	65.0	455	0.1865
No log	66.0	462	0.1877
No log	67.0	469	0.1887
No log	68.0	476	0.1893
No log	69.0	483	0.1893
No log	70.0	490	0.1896
No log	71.0	497	0.1898
0.6248	72.0	504	0.1906
0.6248	73.0	511	0.1910
0.6248	74.0	518	0.1915
0.6248	75.0	525	0.1920
0.6248	76.0	532	0.1924
0.6248	77.0	539	0.1926
0.6248	78.0	546	0.1923
0.6248	79.0	553	0.1924
0.6248	80.0	560	0.1926
0.6248	81.0	567	0.1927
0.6248	82.0	574	0.1928
0.6248	83.0	581	0.1930
0.6248	84.0	588	0.1930
0.6248	85.0	595	0.1929
0.6248	86.0	602	0.1930
0.6248	87.0	609	0.1930
0.6248	88.0	616	0.1933
0.6248	89.0	623	0.1936
0.6248	90.0	630	0.1938
0.6248	91.0	637	0.1940
0.6248	92.0	644	0.1943
0.6248	93.0	651	0.1945
0.6248	94.0	658	0.1945
0.6248	95.0	665	0.1945
0.6248	96.0	672	0.1946
0.6248	97.0	679	0.1945
0.6248	98.0	686	0.1945
0.6248	99.0	693	0.1945
0.6248	100.0	700	0.1945

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.0
• Tokenizers 0.21.0