TAO JIANG

Results 5 issues of TAO JIANG

`tokenizer = get_chat_template( tokenizer, chat_template = "chatml", # Supports zephyr, chatml, mistral, llama, alpaca, vicuna, vicuna_old, unsloth mapping = {"role" : "from", "content" : "value", "user" : "human", "assistant" :...

fixed - pending confirmation

Hello, after I successfully fine-tuned, I run to the last step and when generating text, the following error occurs. May I ask what is the reason? ![image](https://user-images.githubusercontent.com/7599454/224216132-416b7044-dee1-4401-a7c7-c2f76ea37660.png) ![image](https://user-images.githubusercontent.com/7599454/224216173-81b75b67-79d0-4b6f-a654-2cf500ea07d5.png)

bug
good first issue
help wanted

### Motivation Motivation The min_p sampling parameter is becoming quite popular. It's conceptually simple and "makes sense", and (at least anecdotally, according to opinions of many model fine-tuners and users...

This error occurs with some prompts ``` File "/root/flux-api/flux_emphasis.py", line 364, in get_weighted_text_embeddings_flux prompt_tokens_t5_grouped, prompt_weights_t5_grouped = group_tokens_and_weights( ^^^^^^^^^^^^^^^^^^^^^^^^^ File "/root/flux-api/flux_emphasis.py", line 226, in group_tokens_and_weights new_token_ids.append(temp_77_token_ids) ^^^^^^^^^^^^^^^^^ UnboundLocalError: cannot access local...

[https://github.com/ToTheBeginning/PuLID](https://github.com/ToTheBeginning/PuLID) ![image](https://github.com/user-attachments/assets/4a7ea66c-d3b4-4dd7-9226-ca4fbbc5e9c4)