-
Thank you for your work. I am impressed by the significant improvement demonstrated in the Ablation Study. As mentioned in the paper, you used the same model and dataset, but the training strategy app…
-
Let's integrate the unification of each step as "low-hanging fruits" on the "lattice of the Tree of Knowledge" in the "Garden of Eden," where "genetic algorithms mutate ideas." This will symbolize the…
-
本文是对 LLM 进行微调以及二次预训练时，阅读数据训练相关文章的一些学习笔记，包含极少的经验总结。
参考资料：
- [Streamlining Data Preparation for Fine tuning of Large Language Models](https://blog.h2o.ai/blog/streamlining-data-preparation-for-fin…
-
Currently the DPOTrainer is incredibly convenient for fine tuning llms on language preference datasets but no support exists right now for using this class to train multimodal chatbots like LLaVA. In …
-
I have fine-tuned the model using Lora, the config is available here: "Lukee4/biogpt-2020_2labels"
I used BioGPTforSequenceClassification and the fine-tuning worked fine, the results on the test d…
-
I prepared the csv, but in the next step I get this:
```
❯ python -m finetuning --dataset "custom_dataset" --custom_dataset.file "scripts/custom_dataset.py" --whatsapp_username "Jorge"
Tracebac…
-
### Describe the feature request
**Here is the proposal:**
If we have an ONNX file of float16s. This might be a file of 10GB.
Let's say we "quantize" these into int8's. So now the file is of si…
-
## Feature Request
**What is the problem you're currently running into?**
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
We don't yet have mistral mo…
-
### News
- Conference 소식
- [CHI 2023](https://chi2023.acm.org/): 독일 함부르크, 4.23 - 28
- [ICLR 2023](https://iclr.cc/): 르완다 키갈리(Aㅏ), 5.1-5
- Google Deepmind!!!
- Google Brain 과 Deepmind가 하나의 팀…
-
非常感谢您的工作！我在使用DPO训练全量微调后的InternVL2-8B模型遇到了如下问题：
下面是我的微调脚本：
```
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \
swift rlhf \
--rlhf_type dpo \
--model_type internvl2-8b \
--model_id_or_path…