Gemma.cpp hangs on a Gemma 7B model that was finetuned using huggingface peft(QLoRA)
- Lingua principale
- C++
- Stelle
- 7k
- Fork
- 660
- Merge medio
- 20h 43m
- PR unite (30g)
- 33
Descrizione
Hi, thanks for the interesting project!
I create Gemma 7B based model [webbigdata/C3TR-Adapter](https://huggingface.co/webbigdata/C3TR-Adapter).
This model is Huggingface transformer format and translation-only model with original prompt templates fine-tuned by QLoRA.
So, I convert this to pytorch(.ckpt), and result is [f32_merge_model.ckpt](https://huggingface.co/dahara1/gemma_cpp_test/blob/main/f32_merge_model.ckpt)
I have confirmed that f32_merge_model.ckpt works.
Then, run this command, no error message.
python3 convert_weights.py --tokenizer [tokenizer.model](https://huggingface.co/dahara1/gemma_cpp_test/blob/main/tokenizer.model) --weight f32_merge_model.ckpt --output_file [gemma_cpp_merge.bin](https://huggingface.co/dahara1/gemma_cpp_test/blob/main/gemma_cpp_merge.bin) --model_type 7b
Then, run this command, no error message.
./build/compress_weights --weights util/gemma_cpp_merge.bin --model 7b-pt --compressed_weights util/[gemma_cpp_merge.sbs](https://huggingface.co/dahara1/gemma_cpp_test/blob/main/gemma_cpp_merge.bin)
Then, run gemma.cpp, no error message.
./build/gemma --tokenizer util/tokenizer.model --compressed_weights util/gemma_cpp_merge.sbs --model 7b-pt
and input my prompt
[### Instruction:\nTranslate English to Japanese.\n\n### Input:\nThis is a test input.\n\n### Response:\n]
but model can't output anything.
Is there something wrong with the procedure?


Guida per i contributori
Apri la guida per i contributori
Valutazione
Questa issue non è ancora stata valutata.