McGill-NLP llm2vec issues

McGill-NLP / llm2vec

Code for 'LLM2Vec: Large Language Models Are Secretly Powerful Text Encoders'

https://mcgill-nlp.github.io/llm2vec/

MIT License

811 stars 60 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Test llama and Mistral on mteb benchmark

#111 NouamaneELGueddarii opened 17 hours ago
0
Learning implications for loss_scale

#110 daegonYu opened 4 days ago
0
Fine Tuning Script on Custom Data

#109 WorldHellow opened 5 days ago
0
fix _convert_to_str function when instruction is empty

#108 bzantium closed 4 days ago
0
fix `_convert_to_str` to avoid tokenization issue

#107 bzantium closed 4 days ago
0
Does performance increase when learning more steps?

#106 daegonYu opened 6 days ago
0
Some words can not be encoded (upper case and lower case)

#105 wyhwhy closed 4 days ago
4
Instruction for corpus and query on MTEP Evaluation

#104 bzantium opened 1 week ago
1
prepare_for_tokenization (Instruct template) is used for supervised training but not for inference (example)

#103 bzantium closed 1 week ago
0
gemma model

#102 vaibhavad closed 1 week ago
3
Add comment about embeddings device

#101 VProv closed 1 week ago
0
Is it a desired behaviour that _encode always returns embeddings on cpu, even though we pass device argument to it?

#100 VProv closed 6 days ago
3
Bypass wandb check

#99 jeffreyzhang92 closed 6 days ago
2
Usage for multiple Contexts

#98 harshg99 closed 6 days ago
2
Inference example for mask infilling and word prediction?

#97 Mem2019 closed 2 weeks ago
1
Training time quite high

#96 sandeep-krutrim closed 6 days ago
2
What are these sentence preprocessing used for?

#95 ShengYun-Peng closed 6 days ago
2
How to use multiple GPUs

#94 motefly closed 3 weeks ago
1
How do I run MNTP training locally

#93 Georgepitt opened 3 weeks ago
7
Prevent splitting of ModifiedMistralDecoderLayers

#92 hatzel closed 3 weeks ago
4
AttributeError: Can't pickle local object 'add_hook_to_module.<locals>.new_forward'

#91 laughinghugs closed 3 weeks ago
2
OSError when loading the model

#90 JavierCastellD closed 2 weeks ago
2
How to get sentence embedding from last hidden state?

#89 InfAGI closed 3 weeks ago
2
Inquiry Regarding Instruction Addition to Query Statements

#88 shrijayan closed 1 month ago
0
TypeError: LlamaBiModel._update_causal_mask() takes from 4 to 5 positional arguments but 6 were given

#87 guanchangge closed 1 month ago
2
Does the instruction before question have any significance?

#86 aldrinjenson closed 1 month ago
2
Curious about how to directly run LLM for embedding

#85 Georgepitt closed 1 month ago
3
RuntimeError: CUDA error: invalid device ordinal, Compile with TORCH_USE_CUDA_DSA to enable device-side assertions

#84 Hippo88902 closed 1 month ago
6
Sentence pair classification?

#83 davedgd closed 1 month ago
3
Error in _update_causal_mask while running the example code

#82 aldrinjenson closed 1 month ago
4
Enhancement: Include Code for Converting Any LLM to an Encoding Model or Provide Phi-3 Model Support

#81 Shiva-OC closed 1 month ago
2
some special words can not be encoded

#80 Yan2266336 opened 1 month ago
12
eager / sdpa attention

#79 vince62s closed 1 month ago
17
trainning for word task on a custom datatset for ner

#78 SGidentification closed 1 month ago
2
Evaluation on MTEB benchmark

#77 ylwangy opened 1 month ago
6
MNTP learning rate

#76 spookyQubit closed 1 month ago
2
remove condition which returns true for batch size 1

#75 vaibhavad closed 1 month ago
0
For Llama models, bidirectional connections are not enabled when batch size is 1 or no padding token in batch

#74 vaibhavad closed 1 month ago
0
MNTP Question

#73 bdytx5 closed 1 month ago
2
What is the purpose of split text with `!@#$%^&*()`?

#72 fahadh4ilyas closed 1 month ago
2
About Unsupervised contrastive training (SimCSE)

#71 hexi01 closed 1 month ago
1
If I want to use another LLM model, which parts of the code do I need to customize?

#70 liuweie closed 1 month ago
5
Issue when loading model on multiple gpus

#69 614TChen closed 1 month ago
2
Different embeddings obtained when running with different batch size

#68 wufeim closed 1 month ago
2
simcse readme update

#67 vaibhavad closed 1 month ago
0
About MNTP task

#66 Hanser14Forever closed 1 month ago
4
Encoding Device

#65 sramshetty closed 1 month ago
2
Is it possible to load `LLM2Vec` config?

#64 xiaoyuqian2 closed 1 month ago
2
How are the base model weights loaded into llm2vec encoder model?

#63 xiaoyuqian2 closed 1 month ago
2
Simcse training

#62 vaibhavad closed 1 month ago
0