ruadapt_qwen2.5_3B_darulm_bpe_128000_full_lr2e4_bs256

This model is a fine-tuned version of RefalMachine/ruadapt_qwen2.5_3B_darulm_bpe_128000_mean_init on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.7698
• Accuracy: 0.4706

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: 0.0002
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 64
• gradient_accumulation_steps: 2
• total_train_batch_size: 256
• total_eval_batch_size: 128
• optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-05
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.0	1	6.8132	0.1568
2.9865	0.19	2000	2.8235	0.4652
2.9245	0.37	4000	2.7806	0.4695
2.9293	0.56	6000	2.7717	0.4704
2.9299	0.75	8000	2.7698	0.4705
2.9176	0.94	10000	2.7698	0.4706

Framework versions

• Transformers 4.37.2
• Pytorch 2.3.0a0+6ddf5cf85e.nv24.04
• Datasets 2.18.0
• Tokenizers 0.15.2