SmolLM_1_7B_Instruct_qlora_nf4-plaba

This model is a fine-tuned version of HuggingFaceTB/SmolLM-1.7B-Instruct on the generator dataset. It achieves the following results on the evaluation set:

• Loss: 1.7491

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: 4
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• num_devices: 16
• 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
No log	0.8	1	1.9677
No log	1.6	2	1.9588
No log	2.4	3	1.9242
No log	4.0	5	1.8088
No log	4.8	6	1.7755
No log	5.6	7	1.7593
No log	6.4	8	1.7526
1.8621	8.0	10	1.7491

Framework versions

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