llama3.1-8b-ins-lora-100-toy-meta-3

This model is a fine-tuned version of meta-llama/Llama-3.1-8B-Instruct on the metacog0/mbpp_finetune_training_100_qa_True_test_True_code_True_rule_True_100.jsonl dataset. It achieves the following results on the evaluation set:

• Loss: 0.0244

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: 0.0002
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 2
• total_train_batch_size: 128
• total_eval_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss
0.4404	1.0	38	0.4830
0.1329	2.0	76	0.1748
0.0349	3.0	114	0.0600
0.0117	4.0	152	0.0292
0.0025	5.0	190	0.0244

Framework versions

• PEFT 0.13.2
• Transformers 4.45.0.dev0
• Pytorch 2.4.0+cu121
• Datasets 4.0.0
• Tokenizers 0.19.1