Model save

README.md

@@ -0,0 +1,73 @@
+---
+library_name: peft
+tags:
+- trl
+- dpo
+- generated_from_trainer
+base_model: Weni/ZeroShot-3.3.34-Mistral-7b-Multilanguage-3.3.0-merged
+model-index:
+- name: ZeroShot-3.4.8-Mistral-7b-DPO-1.0.0
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# ZeroShot-3.4.8-Mistral-7b-DPO-1.0.0
+
+This model is a fine-tuned version of [Weni/ZeroShot-3.3.34-Mistral-7b-Multilanguage-3.3.0-merged](https://huggingface.co/Weni/ZeroShot-3.3.34-Mistral-7b-Multilanguage-3.3.0-merged) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.2422
+- Rewards/chosen: 2.5475
+- Rewards/rejected: -3.3325
+- Rewards/accuracies: 0.8896
+- Rewards/margins: 5.8800
+- Logps/rejected: -24.9104
+- Logps/chosen: -17.0182
+- Logits/rejected: -1.2365
+- Logits/chosen: -1.2564
+
+## 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: 0.0002
+- train_batch_size: 8
+- eval_batch_size: 16
+- seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- training_steps: 144
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
+|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
+| 0.3541        | 1.02  | 50   | 0.2621          | 1.2744         | -4.9426          | 0.8839             | 6.2170          | -28.1304       | -19.5643     | -1.1344         | -1.1516       |
+| 0.1798        | 2.04  | 100  | 0.2422          | 2.5475         | -3.3325          | 0.8896             | 5.8800          | -24.9104       | -17.0182     | -1.2365         | -1.2564       |
+
+
+### Framework versions
+
+- PEFT 0.8.2
+- Transformers 4.38.2
+- Pytorch 2.1.0+cu118
+- Datasets 2.17.1
+- Tokenizers 0.15.2