tinyllama-1.1b-sft-test3

This model is a fine-tuned version of TinyLlama/TinyLlama-1.1B-Chat-v1.0 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2259

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: 0.0003
• train_batch_size: 16
• eval_batch_size: 10
• seed: 4234
• optimizer: Use OptimizerNames.PAGED_ADAMW 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: 0.03
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
0.3312	0.0778	20	0.2974
0.3523	0.1556	40	0.2538
0.2853	0.2335	60	0.2694
0.2953	0.3113	80	0.2519
0.3104	0.3891	100	0.2379
0.3664	0.4669	120	0.2325
0.3484	0.5447	140	0.2302
0.2604	0.6226	160	0.2322
0.3282	0.7004	180	0.2232
0.2788	0.7782	200	0.2250
0.2589	0.8560	220	0.2260
0.2551	0.9339	240	0.2259

Framework versions

• PEFT 0.15.2
• Transformers 4.51.3
• Pytorch 2.7.0+cu128
• Datasets 3.5.0
• Tokenizers 0.21.1