llama-fine-tuned

This model is a fine-tuned version of meta-llama/Llama-3.2-1B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3503

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: 8
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
1.6035	0.0357	50	0.9702
0.6392	0.0714	100	0.6272
0.5979	0.1071	150	0.5398
0.5629	0.1429	200	0.5044
0.4761	0.1786	250	0.4689
0.4998	0.2143	300	0.4494
0.4363	0.25	350	0.4524
0.4433	0.2857	400	0.4322
0.4882	0.3214	450	0.4135
0.4316	0.3571	500	0.4017
0.389	0.3929	550	0.3951
0.4041	0.4286	600	0.3908
0.456	0.4643	650	0.3860
0.3872	0.5	700	0.3788
0.3962	0.5357	750	0.3792
0.3524	0.5714	800	0.3762
0.3409	0.6071	850	0.3700
0.421	0.6429	900	0.3746
0.349	0.6786	950	0.3634
0.4194	0.7143	1000	0.3665
0.3621	0.75	1050	0.3607
0.3663	0.7857	1100	0.3603
0.3434	0.8214	1150	0.3592
0.3609	0.8571	1200	0.3553
0.342	0.8929	1250	0.3524
0.3889	0.9286	1300	0.3513
0.3604	0.9643	1350	0.3508
0.354	1.0	1400	0.3503

Framework versions

• PEFT 0.13.2
• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Tokenizers 0.21.0