LMOps

LMOps copied to clipboard

Bumps [black](https://github.com/psf/black) from 22.1.0 to 24.3.0. Release notes Sourced from black's releases. 24.3.0 Highlights This release is a milestone: it fixes Black's first CVE security vulnerability. If you run Black...

dependabot[bot]

The update method in the UCB algorithm is inconsistent with the paper and code

2

Q(p) for each prompt in the UCB algorithm of the paper is updated to Q(p) + r/N(p), ![Uploading 2024331203750.jpg…]() The following table describes the project update code ## def update(self,...

kerala21

Hi, thanks for your great job. I run the `download_data.sh` script and obtain the dataset sucessfully, but I have some questions about what exactly each file means: 1. What is...

OStars

ModuleNotFoundError: No module named 'deepspeed'

1

When i run the following command, the problem with the title arises ``` bash /home/iotsc01/xinpengq/LMOps-main/minillm/scripts/llama2/sft/sft_7B.sh /home/iotsc01/LMOps-main/minillm ``` the scripts file is following: ``` #! /bin/bash BASE_MODEL=/home/iotsc01/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-hf/snapshots/8cca527612d856d7d32bd94f8103728d614eb852 PYTHON_ENV_PATH="/home/iotsc01/anaconda3/envs/distil/bin/python" MASTER_ADDR=localhost MASTER_PORT=${2-2012} NNODES=1...

qxpBlog

prompt_optimization

2

The dataset for prompt optimization should consist of four, but the source code only provides two. Is there a missing part?

chensimian

Missing Jailbreak dataset from protegi?

2

Hi, in the protegi paper ("Automatic Prompt Optimization with “Gradient Descent” and Beam Search" 2023) the authors stated that the method was evaluated across 4 NLP benchmark tasks: Jailbreak, Ethos,...

tboen1

[tuna] Libraries are conflicting and/or very aged

5

So disappointed of what is released here. these are just non working pieces. Funny that in train.py for example you have: from custom import CustomTrainer, but custom is actually have...

batawfic

top-p < 1 fails inf assertion

1

When I set top-p in the arguments to a value less than 1, I have [this assert](https://github.com/microsoft/LMOps/blob/main/minillm/minillm/utils.py#L57) failing. Which seems to make sense since [this line puts them in](https://github.com/microsoft/LMOps/blob/main/minillm/minillm/utils.py#L44). What's...

artsobolev

Questions about the free-law data used in the paper "Adapt LLM to domains"

2

Dear Authors, you have undoubtedly done an excellent job (domain-specific post-pre-training). But I have a small question about the size of the free-law data used in the original paper, I...

WUHU-G

Hi. Can I ask about the query of GPT-4 evaluation in detail? I tried to """ We would like to request your feedback on the performance of two AI assistants...

jongwooko