Dinesh Khandelwal

Hi, Can you please tell me what is the difference between fc_8 features and crf features. There are different trained models some has fc_8 in the test.prototxt some has crf....

Thanks @brozi for the reply. Few more clarifications questions. 1. In case we don't have unit tests, then in AE and BT training do we require to create 'train.cpp_sa-java_sa.java_sa.pth', val.cpp_sa-java_sa.java_sa.pth,...

Thanks @brozi. Can you please tell me how MT and BT steps are training using the cross-entropy loss ? Is it summing binary cross-entropy over all the tokens in the...

@geronimi73 sorry for late reply. Below is the code: ``` from peft import PeftModel, PeftConfig from transformers import AutoModelForCausalLM import torch from datasets import load_dataset from transformers import AutoTokenizer import...

@geronimi73 I have updated the code above, please check now Here is one checkpoint: [falcon-7b-instruct_ckpt_step_4000.zip](https://github.com/huggingface/accelerate/files/14121322/falcon-7b-instruct_ckpt_step_4000.zip) Also, without using accelerate I am able to run inference on single GPU.

@geronimi73 sorry for incorrect statement above (I have corrected it). I am able to do inference with prompt tuned 7B model on single gpu without accelerate.

@geronimi73 any pointers on what is the issue here. I have tried loading the base model (`tiiuae/falcon-7b-instruct`) without any PEFT adapter that worked. I have loaded the base model using...

@pacman100 here is the minimal reproducible example. ``` from peft import PeftModel, PeftConfig from transformers import AutoModelForCausalLM import torch from absl import flags from absl import app from accelerate import...

@geronimi73 **Command :** `accelerate launch slot_filling_inference_accelerate_demo.py --ckpt_path ./test_peft/falcon-7b-instruct_ckpt_step_4000` **Output:** ``` The following values were not passed to `accelerate launch` and had defaults used instead: `--num_processes` was set to a value...

@geronimi73 getting the same error with above code as well.