metadata
base_model: seanmemery/MLP-FinLLM-7b-it
tags:
- trl
- dpo
- unsloth
- generated_from_trainer
model-index:
- name: MLP-FinLLM-dpo-7b
results: []
MLP-FinLLM-dpo-7b
This model is a fine-tuned version of seanmemery/MLP-FinLLM-7b-it on the None dataset. It achieves the following results on the evaluation set:
- Loss: 11.2447
- Rewards/chosen: -124.0
- Rewards/rejected: -119.0
- Rewards/accuracies: 0.5207
- Rewards/margins: -5.125
- Logps/rejected: -1216.0
- Logps/chosen: -1280.0
- Logits/rejected: -5.5625
- Logits/chosen: -5.5625
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: 0.003
- train_batch_size: 32
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 2
Training results
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 27.875 | 0.69 | 50 | 33.2248 | -154.0 | -137.0 | 0.5041 | -16.625 | -1392.0 | -1576.0 | -7.9062 | -7.9062 |
| 14.375 | 1.39 | 100 | 11.2447 | -124.0 | -119.0 | 0.5207 | -5.125 | -1216.0 | -1280.0 | -5.5625 | -5.5625 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.0
- Datasets 2.18.0
- Tokenizers 0.15.2