CodePhi-3-mini-4k-instruct-APPS

This model is a fine-tuned version of microsoft/Phi-3-mini-128k-instruct on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7651

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: 5e-06
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• training_steps: 1200

Training results

Training Loss	Epoch	Step	Validation Loss
0.8306	0.0833	100	0.8445
0.6325	0.1667	200	0.8164
0.6696	0.25	300	0.7934
0.643	0.3333	400	0.7815
0.778	0.4167	500	0.7737
0.7273	0.5	600	0.7696
0.712	0.5833	700	0.7676
0.6477	0.6667	800	0.7658
0.6245	0.75	900	0.7652
0.6959	0.8333	1000	0.7650
0.7236	0.9167	1100	0.7650
0.6774	1.0	1200	0.7651

Framework versions

• PEFT 0.11.0
• Transformers 4.40.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1