metadata
library_name: transformers
license: other
base_model: hon9kon9ize/Qwen2.5-7B-cpt
tags:
- llama-factory
- full
- generated_from_trainer
model-index:
- name: CantoneseLLMChat-v1.0
results: []
CantoneseLLMChat-v1.0
This model is a fine-tuned version of hon9kon9ize/Qwen2.5-7B-cpt on the dpo_v1 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3170
- Rewards/chosen: -0.7307
- Rewards/rejected: -3.1239
- Rewards/accuracies: 0.8464
- Rewards/margins: 2.3931
- Logps/rejected: -226.0627
- Logps/chosen: -191.7517
- Logits/rejected: -1.5777
- Logits/chosen: -1.5363
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.0835 | 1.3495 | 500 | 0.3282 | -0.6181 | -2.8043 | 0.8494 | 2.1862 | -222.8677 | -190.6253 | -1.5563 | -1.5215 |
| 0.1151 | 2.6991 | 1000 | 0.3186 | -0.7277 | -3.1230 | 0.8524 | 2.3953 | -226.0546 | -191.7211 | -1.5777 | -1.5363 |
Framework versions
- Transformers 4.45.0
- Pytorch 2.4.1+cu121
- Datasets 2.20.0
- Tokenizers 0.20.0