qwen25_math_eng_rm

This model is a fine-tuned version of Jennny/qwen25_7b_sft_eng_math on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.8559
  • Accuracy: 0.7567

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: 2
  • eval_batch_size: 2
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 8
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 128
  • total_eval_batch_size: 16
  • optimizer: Use OptimizerNames.PAGED_ADAMW with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.03
  • num_epochs: 2

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.2222 0.1677 10 1.4139 0.5333
0.065 0.3354 20 0.9037 0.6067
0.0552 0.5031 30 0.7860 0.68
0.0371 0.6709 40 1.0758 0.7133
0.0731 0.8386 50 0.7435 0.7433
0.0534 1.0 60 0.6848 0.7467
0.0035 1.1677 70 1.2106 0.7433
0.0064 1.3354 80 1.2054 0.7467
0.0223 1.5031 90 1.3114 0.7367
0.0503 1.6709 100 0.9685 0.7467
0.0387 1.8386 110 0.8559 0.7567

Framework versions

  • Transformers 4.49.0
  • Pytorch 2.5.1+cu124
  • Datasets 3.3.2
  • Tokenizers 0.21.1
Downloads last month
-
Safetensors
Model size
7B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Jennny/qwen25_math_eng_rm

Base model

Qwen/Qwen2.5-7B
Finetuned
(6)
this model