norallm_normistral-7b-warm-instruct_4bit_quant

This model is a fine-tuned version of norallm/normistral-7b-warm-instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0707

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: 0.0005
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 32
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.7	0.1954	100	0.5815
0.371	0.3907	200	0.3392
0.2114	0.5861	300	0.2257
0.1371	0.7814	400	0.1624
0.1141	0.9768	500	0.1242
0.0852	1.1719	600	0.1016
0.0756	1.3673	700	0.0852
0.0586	1.5626	800	0.0758
0.0525	1.7580	900	0.0707

Framework versions

• PEFT 0.15.2
• Transformers 4.51.3
• Pytorch 2.7.0+cu118
• Datasets 3.5.1
• Tokenizers 0.21.1