LoRA · Issues· 114 open

Feedback on Training Parameters for Fine-Tuning a 7B RAG LLM, slow training:/

Need Help with Fine-Tuning LoRa Using Diffusion Pipe for Hyperrealistic Object Videos

Why not merge BA during training?

Reproduce Lora results is close but not accurate

Cāsnālā

تمکن مالی فرانسه

ValueError: Target module CLIPEncoderLayer，only the following modules are supported: `torch.nn.Linear`, `torch.nn.Embedding`, `torch.nn.Conv2d`, `transformers.pytorch_utils.Conv1D`.

Live clock

To wanna take job internship

Can not reproduce the result of Roberta-Base

Confusing logic for `reset_parameters()` in `Embedding` and `Linear` layer

`lora.Linear.weight` Parameters Change After Loading Checkpoint in Train Mode Leading to Inconsistent Evaluation Results

Layers.py not being executed

Entire model weights saved after fine-tuning instead of only the LoRA weights?

Questions about replicating NLU experiments