Use this tag for questions about large language models (LLM), trained deep-learning artificial intelligence algorithms that interpret and generate natural language text.
Questions tagged [large-language-model]
118 questions
0
votes
0 answers
How to use results in output directory in producing environment?
I have trained facebook/opt-6.7b with [BELLE][1]'s dataset, however the biggest file optimizer.pt in qlora/output/checkpoint is only 1.1GB, and size of the whole facebook/opt-6.7b is about 13GB.
so where is the real model and how to use it in…
forestbat
- 453
- 4
- 10
0
votes
1 answer
langchain + Weaviate how to access multiple columns at once
I have created a schema with multiple properties in Weaviate. using the following approach:
for row in tqdm(data, total=len(data)):
client.data_object.create(data_object=row, class_name=INDEX_NAME)
here is a sample of the data (1…
Clem
- 51
- 7
0
votes
1 answer
What is the license of sentence-transformers/multi-qa-mpnet-base-dot-v1? Is it Apache 2.0 or MIT?
what is the license under which sentence-transformers/multi-qa-mpnet-base-dot-v1 can be used? Apache 2.0 or MIT or ? where can we find more information on this.
ashwar hussain
- 115
- 1
- 9
0
votes
0 answers
Can Databricks Dolly be trained on Databricks tables for generating insights using prompts?
I'm exploring the capabilities of Databricks specifically Databricks Dolly, and I'm wondering if it's possible to train Dolly on Databricks tables to generate insights by writing prompts.
I have a Databricks environment set up and I'm working with…
nilesh1212
- 1,561
- 2
- 26
- 60
0
votes
1 answer
How to get the embeddings from the first 4 layers of pre-trained LLMs such as CodeBERT/GraphCodeBERT?
I need to get the embeddings from a pre-trained LLM. As of now I am doing something like this:
def gen_embeddings(self,code):
tokenized_input_pos = self.tokenizer(code, return_tensors="pt", padding=True, truncation=True)
with…
not-a-bot
- 25
- 3
0
votes
0 answers
Meet bug when using Llama-CUDA error: device-side assert triggered
enter image description here
I fine-tune the Llama-7b-hf model on downstream task and use eos_token as the pad_token
When evaluating using model.generate(), this error occurs at 5th batch (the last 4 batches run without any trouble). I print…
keheng wang
- 1
- 1
0
votes
0 answers
Starcoder - Why NVIDIA Tesla T4 GPU switching is not happening causing OutOfMemoryError?
In order to fine tune Starcoder LLM model on my GCP instance, I have setup 4 NVIDIA Tesla T4 GPUs (16GB each)
I installed nvitop to monitor the usage of the GPUs while finetuning.
I have also installed the CUDA toolkit on the VM. (checked if it's…
Aadesh
- 403
- 3
- 13
0
votes
1 answer
How can I use/load the downloaded Hugging Face models from snapshot_download?
I have downloaded the model from Hugging Face using snapshot_download, e.g.,
from huggingface_hub import snapshot_download
snapshot_download(repo_id="facebook/nllb-200-distilled-600M", cache_dir="./")
And when I list the directory, I see:
ls…
alvas
- 115,346
- 109
- 446
- 738
0
votes
1 answer
Internal error encountered | DATA_FETCHING_EXCEPTION | Vertex AI Language, Tune a model
I have been trying to tune a model for a week now using the Vertex AI - Language, Tune a model following this tutorial.
I have used the the sample JSONL as mentioned in the link:
{"input_text": "question: How many people live in Beijing? context:…
Mohit Arvind khakharia
- 397
- 4
- 11
0
votes
0 answers
Match reviews with concepts
I have "N" product reviews and "K" concepts such as "Color", "Audio", "User Experience", "Durability", etc. I want to get back an "N x K" boolean numpy array where 1 indicates if the review talks about the said concept, and 0 indicates it does not.…
Shrayani Mondal
- 160
- 1
- 7
0
votes
0 answers
fine tune gpt3 with tabular data
As the title says, how would one fine tune a LLM with tabular data? My initial sense is that LLM are not suited to learn from tabular data unless the tabular data is restructured into grammatical form, and even then, I have reservations about this…
darrahts
- 365
- 1
- 10
0
votes
0 answers
Low Frequency Tokens in BPE
Learning about tokenization, I implemented the BPE algorithm and trained it on a small corpus: the full text of Harry Potter. I noticed the following thing: my vocabulary contains tokens for "Dumbledore" and " Dumbledore" (notice the leading space),…
Yo.
- 15
- 6
0
votes
0 answers
Cmake --build error, GPT4ALL , Cannot open include file: 'unistd.h'
I am trying to run gpt4all model on my windows 10, but when i run this command 'cmake --build . --parallel' I got this error
Cannot open include file: 'unistd.h': No such file or directory…
GEEK901
- 1
- 1
0
votes
1 answer
How to split voices per bar in ABC music notation
Background
I've recently been trying to compose music, but I'm an absolute novice and I've no background in music theory. So I wanted to train LLaMA to help me give suggestions on how to continue a piece of music with multiple instruments. However,…
Coen Hacking
- 13
- 1
- 7
0
votes
1 answer
How to use LLMChain with llm model stored in disk
I need to use LLMChain with locally stored model. I have below code.
llm_chain = LLMChain(prompt=prompt, llm = HuggingFaceHub(repo_id="google/flan-t5-large", model_kwargs={..some params}))
Instead of repo_id , I need to provide local path. Please…
Khushi
- 325
- 1
- 11
- 32