lmind_hotpot_train8000_eval7405_v1_recite_qa_meta-llama_Llama-2-7b-hf_lora2

This model is a fine-tuned version of meta-llama/Llama-2-7b-hf on the tyzhu/lmind_hotpot_train8000_eval7405_v1_recite_qa dataset. It achieves the following results on the evaluation set:

• Loss: 1.7506
• Accuracy: 0.6631

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: 2
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 4
• 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: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.0606	1.0	1089	0.9943	0.7197
0.9709	2.0	2178	0.9236	0.7259
0.8688	3.0	3267	0.8481	0.7329
0.9838	4.0	4357	0.9606	0.7221
3.857	5.0	5446	10.4449	0.4496
0.9367	6.0	6535	0.9551	0.7223
0.8797	7.0	7624	0.9217	0.7259
0.9011	8.0	8714	0.9460	0.7236
0.8435	9.0	9803	0.9380	0.7243
1.9697	10.0	10890	1.7506	0.6631

Framework versions

• Transformers 4.34.0
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.14.1