ob-10k-30percent

This model is a fine-tuned version of Qwen/Qwen2.5-7B-Instruct on the ob-10k-30percent dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1617

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: 1e-05
  • train_batch_size: 1
  • eval_batch_size: 1
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 4
  • total_train_batch_size: 4
  • total_eval_batch_size: 4
  • optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: cosine
  • num_epochs: 1

Training results

Training Loss Epoch Step Validation Loss
0.2535 0.0553 100 0.2181
0.2985 0.1106 200 0.2109
0.1578 0.1659 300 0.2008
0.1775 0.2212 400 0.2058
0.1516 0.2765 500 0.2057
0.2426 0.3319 600 0.1940
0.1668 0.3872 700 0.1940
0.1783 0.4425 800 0.1869
0.1702 0.4978 900 0.1793
0.2285 0.5531 1000 0.1762
0.1273 0.6084 1100 0.1766
0.1044 0.6637 1200 0.1706
0.1577 0.7190 1300 0.1640
0.1452 0.7743 1400 0.1632
0.1408 0.8296 1500 0.1635
0.1579 0.8850 1600 0.1623
0.1372 0.9403 1700 0.1624
0.1433 0.9956 1800 0.1622

Framework versions

  • Transformers 4.46.1
  • Pytorch 2.4.0+cu121
  • Datasets 3.1.0
  • Tokenizers 0.20.3
Downloads last month
1
Safetensors
Model size
7.62B params
Tensor type
BF16
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for cutelemonlili/Qwen-7B-instruct-ob-10k-30percent

Base model

Qwen/Qwen2.5-7B
Finetuned
(2449)
this model