Qwen2-0.5B-Reward

This model is a fine-tuned version of Qwen/Qwen2-0.5B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5212
  • Accuracy: 0.731

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: 8
  • eval_batch_size: 8
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 8
  • total_train_batch_size: 64
  • total_eval_batch_size: 64
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 1.0

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.6316 0.0516 50 0.5943 0.666
0.573 0.1032 100 0.5857 0.698
0.5809 0.1548 150 0.5718 0.705
0.5493 0.2064 200 0.5450 0.714
0.5649 0.2580 250 0.5483 0.713
0.5585 0.3096 300 0.5265 0.734
0.5431 0.3612 350 0.5295 0.732
0.5209 0.4128 400 0.5334 0.735
0.5414 0.4644 450 0.5409 0.726
0.525 0.5160 500 0.5387 0.731
0.5242 0.5676 550 0.5255 0.727
0.521 0.6192 600 0.5208 0.727
0.5227 0.6708 650 0.5191 0.736
0.5132 0.7224 700 0.5186 0.728
0.5145 0.7740 750 0.5236 0.729
0.514 0.8256 800 0.5249 0.728
0.5087 0.8772 850 0.5261 0.725
0.5009 0.9288 900 0.5229 0.727
0.4989 0.9804 950 0.5213 0.731

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0+cu121
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
28
Safetensors
Model size
494M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for lewtun/Qwen2-0.5B-Reward

Base model

Qwen/Qwen2-0.5B
Finetuned
(55)
this model