lmind_nq_train6000_eval6489_v1_recite_qa_v3__home_aiops_zhuty_lm_indexer_data_tyzhu_lmin

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4657
• Accuracy: 0.7995

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: 1
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• total_eval_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 20.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.4253	1.0	529	0.5042	0.7770
0.3444	2.0	1058	0.4371	0.7875
0.2679	3.0	1587	0.3925	0.7946
0.2195	4.0	2116	0.3709	0.7977
0.1889	5.0	2645	0.3616	0.7998
0.1724	6.0	3174	0.3608	0.8002
0.1573	7.0	3703	0.3646	0.8006
0.144	8.0	4232	0.3774	0.8000
0.1353	9.0	4761	0.3889	0.8000
0.1281	10.0	5290	0.3975	0.8000
0.124	11.0	5819	0.4108	0.7998
0.1169	12.0	6348	0.4183	0.8001
0.1128	13.0	6877	0.4249	0.7997
0.1108	14.0	7406	0.4259	0.8004
0.1078	15.0	7935	0.4435	0.7994
0.1065	16.0	8464	0.4421	0.7999
0.104	17.0	8993	0.4450	0.7998
0.103	18.0	9522	0.4554	0.7995
0.1033	19.0	10051	0.4556	0.7997
0.1041	20.0	10580	0.4657	0.7995

Framework versions

• PEFT 0.5.0
• Transformers 4.40.2
• Pytorch 2.3.0
• Datasets 2.19.1
• Tokenizers 0.19.1