learning_rate: 0.0001
train_batch_size: 32
eval_batch_size: 64
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: 30
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu 1 Msl	Bleu 2 Msl	Bleu 3 Msl	Bleu 4 Msl	Ter Msl	Ter Asl
No log	1.0	75	1.8453	0.3	0.0100	0.0034	0.0018	{'score': 554.1841004184101, 'num_edits': 5298, 'ref_length': 956.0}	100
No log	2.0	150	1.2624	0.4033	0.0116	0.0038	0.0020	{'score': 381.4853556485356, 'num_edits': 3647, 'ref_length': 956.0}	100
No log	3.0	225	1.1691	0.4367	0.0121	0.0039	0.0020	{'score': 33.68200836820084, 'num_edits': 322, 'ref_length': 956.0}	100
No log	4.0	300	1.2114	0.42	0.0119	0.0038	0.0020	{'score': 37.238493723849366, 'num_edits': 356, 'ref_length': 956.0}	100
No log	5.0	375	1.2784	0.4167	0.0118	0.0038	0.0020	{'score': 35.66945606694561, 'num_edits': 341, 'ref_length': 956.0}	100
No log	6.0	450	1.2639	0.4267	0.0119	0.0038	0.0020	{'score': 35.46025104602511, 'num_edits': 339, 'ref_length': 956.0}	100
0.5532	7.0	525	1.2715	0.4333	0.0120	0.0039	0.0020	{'score': 35.77405857740586, 'num_edits': 342, 'ref_length': 956.0}	100
0.5532	8.0	600	1.3544	0.4467	0.0122	0.0039	0.0020	{'score': 33.36820083682008, 'num_edits': 319, 'ref_length': 956.0}	100
0.5532	9.0	675	1.3177	0.45	0.0123	0.0039	0.0020	{'score': 32.42677824267782, 'num_edits': 310, 'ref_length': 956.0}	100
0.5532	10.0	750	1.3129	0.4333	0.0120	0.0039	0.0020	{'score': 33.68200836820084, 'num_edits': 322, 'ref_length': 956.0}	100
0.5532	11.0	825	1.3626	0.44	0.0121	0.0039	0.0020	{'score': 32.94979079497908, 'num_edits': 315, 'ref_length': 956.0}	100
0.5532	12.0	900	1.3124	0.4767	0.0126	0.0040	0.0021	{'score': 33.15899581589959, 'num_edits': 317, 'ref_length': 956.0}	100
0.5532	13.0	975	1.3840	0.46	0.0124	0.0039	0.0020	{'score': 33.36820083682008, 'num_edits': 319, 'ref_length': 956.0}	100
0.04	14.0	1050	1.3624	0.47	0.0125	0.0040	0.0021	{'score': 31.799163179916317, 'num_edits': 304, 'ref_length': 956.0}	100
0.04	15.0	1125	1.3695	0.46	0.0124	0.0039	0.0020	{'score': 31.903765690376567, 'num_edits': 305, 'ref_length': 956.0}	100
0.04	16.0	1200	1.3498	0.4333	0.0120	0.0039	0.0020	{'score': 32.11297071129707, 'num_edits': 307, 'ref_length': 956.0}	100
0.04	17.0	1275	1.3573	0.45	0.0123	0.0039	0.0020	{'score': 32.11297071129707, 'num_edits': 307, 'ref_length': 956.0}	100
0.04	18.0	1350	1.4796	0.4533	0.0123	0.0039	0.0020	{'score': 31.903765690376567, 'num_edits': 305, 'ref_length': 956.0}	100
0.04	19.0	1425	1.3966	0.46	0.0124	0.0039	0.0020	{'score': 30.648535564853557, 'num_edits': 293, 'ref_length': 956.0}	100
0.0192	20.0	1500	1.3763	0.4567	0.0124	0.0039	0.0020	{'score': 32.63598326359833, 'num_edits': 312, 'ref_length': 956.0}	100
0.0192	21.0	1575	1.3646	0.4633	0.0124	0.0039	0.0020	{'score': 31.799163179916317, 'num_edits': 304, 'ref_length': 956.0}	100
0.0192	22.0	1650	1.3742	0.4633	0.0124	0.0039	0.0020	{'score': 32.74058577405858, 'num_edits': 313, 'ref_length': 956.0}	100
0.0192	23.0	1725	1.3672	0.48	0.0127	0.0040	0.0021	{'score': 31.903765690376567, 'num_edits': 305, 'ref_length': 956.0}	100
0.0192	24.0	1800	1.3960	0.4533	0.0123	0.0039	0.0020	{'score': 31.799163179916317, 'num_edits': 304, 'ref_length': 956.0}	100
0.0192	25.0	1875	1.3700	0.47	0.0125	0.0040	0.0021	{'score': 33.054393305439326, 'num_edits': 316, 'ref_length': 956.0}	100
0.0192	26.0	1950	1.4044	0.4567	0.0124	0.0039	0.0020	{'score': 32.00836820083682, 'num_edits': 306, 'ref_length': 956.0}	100
0.013	27.0	2025	1.3973	0.46	0.0124	0.0039	0.0020	{'score': 32.53138075313807, 'num_edits': 311, 'ref_length': 956.0}	100
0.013	28.0	2100	1.4086	0.4667	0.0125	0.0040	0.0020	{'score': 31.276150627615063, 'num_edits': 299, 'ref_length': 956.0}	100
0.013	29.0	2175	1.4186	0.4533	0.0123	0.0039	0.0020	{'score': 32.42677824267782, 'num_edits': 310, 'ref_length': 956.0}	100
0.013	30.0	2250	1.4174	0.4567	0.0124	0.0039	0.0020	{'score': 32.21757322175732, 'num_edits': 308, 'ref_length': 956.0}	100

Framework versions

Transformers 4.49.0
Pytorch 2.5.1+cu124
Datasets 3.3.2
Tokenizers 0.21.0

vania2911/exp4_10partition_modeloorig

exp4_10partition_modeloorig

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

vania2911/
exp4_10partition_modeloorig