mistral_7b_0_3-coding-gpt4o-100k2

This model is a fine-tuned version of mistralai/Mistral-7B-v0.3 on the llama-duo/synth_coding_dataset_dedup dataset. It achieves the following results on the evaluation set:

• Loss: 1.1573

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: 4
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 2
• 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: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
0.419	0.9992	640	1.0503
0.3829	2.0	1281	1.0418
0.3774	2.9992	1921	1.0361
0.355	4.0	2562	1.0419
0.3413	4.9992	3202	1.0548
0.318	6.0	3843	1.0761
0.3056	6.9992	4483	1.1023
0.2857	8.0	5124	1.1256
0.2732	8.9992	5764	1.1472
0.2714	9.9922	6400	1.1573

Framework versions

• PEFT 0.12.0
• Transformers 4.44.0
• Pytorch 2.4.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1