smolLM

This model is a fine-tuned version of HuggingFaceTB/SmolLM-360M-Instruct on the generator dataset. It achieves the following results on the evaluation set:

• Loss: 1.8760

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: 0.001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• 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: 0.03
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
2.2721	0.9756	10	2.1262
2.0927	1.9512	20	2.0278
2.0071	2.9268	30	1.9690
1.9512	4.0	41	1.9282
1.9247	4.9756	51	1.9045
1.9024	5.9512	61	1.8897
1.88	6.9268	71	1.8809
1.8788	8.0	82	1.8767
1.8763	8.9756	92	1.8760
1.8735	9.7561	100	1.8760

Framework versions

• PEFT 0.10.0
• Transformers 4.40.0
• Pytorch 2.1.0
• Datasets 2.18.0
• Tokenizers 0.19.1