openchat
openchat copied to clipboard
OpenChat: Advancing Open-source Language Models with Imperfect Data
mixtral-of-experts now released,are you going to support MOE model? https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1
draft of multimodal openchat training, clip + mlp + mistral (llava arch)
After using the google search plugin when I return to the regular LLM it is unable to generate a response I get the following error ``` Exception in callback functools.partial(,...
How do I use the Google plugin? Where can I find the correct API key and CSE ID? I created a set on the google programmable search engine but I...
Here is the SFTtrainer method i used for finetuning mistral ``` trainer = SFTTrainer( model=peft_model, train_dataset=data, peft_config=peft_config, dataset_text_field=" column name", max_seq_length=3000, tokenizer=tokenizer, args=training_arguments, packing=packing, ) trainer.train() ``` I found different...
I am finetuning the mistral model using the following configurations ``` training_arguments = TrainingArguments( output_dir=output_dir, per_device_train_batch_size=per_device_train_batch_size, gradient_accumulation_steps=gradient_accumulation_steps, optim=optim, save_steps=save_steps, logging_strategy="steps", logging_steps=10, learning_rate=learning_rate, weight_decay=weight_decay, fp16=fp16, bf16=bf16, max_grad_norm=max_grad_norm, max_steps=13000, warmup_ratio=warmup_ratio, group_by_length=group_by_length, lr_scheduler_type=lr_scheduler_type...
It seems that OpenChat 3.5 1210 is not open-source and not available on Hugging Face.
Thanks for open-sourcing this! I am trying to follow the instructions for tokenizing the data, but it fails with the stack trace below. I'm just using two lines of dummy...
So the output of my fine-tuned open chat model ends abruptly and I ideally want it to complete the paragraph/sentences/code which it was it between. Although I have provided max_new_tokens...
1. i start it with: python -m ochat.serving.openai_api_server --model openchat/openchat-3.5-1210 --host *.*.*.* 2. then i write a http request ef call_fastapi_service(input_text: str): url = "http://*.*.*.*:18888/v1/chat/completions" data={ "model":"openchat/openchat-3.5-1210", "messages" : [...