graf/Llama-3.2-1B-RM-GSM8k

reward

This model is a fine-tuned version of meta-llama/Llama-3.2-1B-Instruct on the gsm8k_llama3.2-1B_128_1ep dataset. It achieves the following results on the evaluation set:

• Loss: 0.6323
• Accuracy: 0.5876

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: 5e-06
• train_batch_size: 2
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 8
• total_train_batch_size: 128
• total_eval_batch_size: 32
• 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
• lr_scheduler_warmup_ratio: 0.03
• num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7961	0.0855	5	0.5987	0.6209
0.5229	0.1709	10	0.6532	0.5292
0.4513	0.2564	15	0.6738	0.4716
0.5194	0.3419	20	0.6736	0.4769
0.4323	0.4274	25	0.6592	0.5277
0.4733	0.5128	30	0.6529	0.5504
0.3579	0.5983	35	0.6559	0.5299
0.2933	0.6838	40	0.6503	0.5322
0.3119	0.7692	45	0.6438	0.5504
0.3451	0.8547	50	0.6330	0.5883
0.3238	0.9402	55	0.6331	0.5845

Framework versions

• Transformers 4.46.1
• Pytorch 2.5.1
• Datasets 3.1.0
• Tokenizers 0.20.1