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

1.31k stars 95 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Unable to Reproduce Results for Supervised Training on Echo Dataset with Mistral-7B-Instruct-v2

#154 ThonyPan opened 1 week ago
0
Introducing support for the IBM Granite 3.0 Model family.

#153 kailas711 opened 1 month ago
0
NaN in Generated Embeddings

#152 superctj opened 1 month ago
4
Discrepancy in STS17 Results for S-LLaMA-1.3B

#151 newfish-lab opened 1 month ago
3
Issue with mntp training for Llama 3.2 model

#150 sandeep-krutrim opened 1 month ago
0
Issue with mntp training for Llama 3.2 model

#149 sandeep-krutrim closed 1 month ago
0
mteb eval custom script

#148 vaibhavad closed 1 month ago
0
Llama 31 and latest transformer version

#147 vaibhavad closed 1 month ago
0
feat(pooling): faster avg. with EmbeddingBags

#146 fdschmidt93 opened 1 month ago
3
Is it possible to resume from checkpoint in run_mntp.py?

#145 yallk opened 2 months ago
1
Eager/Spda Attention have lower results compared to Flash Attention in simcse stage

#144 ThonyPan opened 2 months ago
2
Where is the code for reproducing "bi+mntp+simcse"?

#143 Sherrylife closed 1 month ago
2
Implement Phi 3 Training

#142 HydrogenBombaklot opened 3 months ago
0
How to use the llama 3.1 to generate the embedding?

#141 Yan2266336 opened 3 months ago
3
MTEB Evaluation Running Time

#140 stefanhgm opened 3 months ago
10
OSError: We couldn't connect to 'https://huggingface.co' to load this file

#139 mooc-jason opened 3 months ago
1
getting error when trying to do quantization using bitsandbytes

#138 sandeep-krutrim opened 3 months ago
0
"The MNTP LoRA weights are merged into the base model, and the trainable LoRA weights are initialized with SimCSE weights."

#137 cultivater closed 1 month ago
2
Add qwen2 config file to mntp

#136 youssefkhalil320 opened 3 months ago
0
Unable to Reproduce LLM2Vec Training Results Using GradCache on Echo Dataset

#135 viet-data opened 3 months ago
10
fixes #128

#134 vaibhavad closed 3 months ago
0
Possible to train Llama 3.1?

#133 mosh98 opened 3 months ago
7
BUG when testing results of MTEB retrieval

#132 tianyumyum opened 3 months ago
4
Finetuning code for sequence classification, NLI task ?

#131 sandeep-krutrim closed 2 months ago
2
Access denied error

#130 Sreelakshmi-k closed 2 months ago
2
Inference on multiple GPUs

#129 fa1c4 opened 3 months ago
0
Loading model + merged adapter is different to model + adapter?

#128 DorotheaMueller closed 2 months ago
6
Implement gemma2 training

#127 bzantium opened 4 months ago
2
Implement Gemma2 training

#126 bzantium opened 4 months ago
0
Sentence/document classification

#125 versae closed 2 months ago
3
Test on german clustering

#124 NouamaneELGueddarii closed 3 months ago
2
Unable to load merged model for MTEB evaluation

#123 sandeep-krutrim opened 4 months ago
3
Failed to reproduce MTEB results

#122 ThonyPan closed 4 months ago
4
Bug in multiprocess?

#121 fa1c4 closed 4 months ago
3
Building with `pretrained_model` not `instruct_model`?

#120 su-park closed 4 months ago
1
MistralBiModel.forward() got an unexpected keyword argument 'cache_position'

#119 su-park closed 4 months ago
2
Mteb eval

#118 vaibhavad closed 4 months ago
0
qwen2 model & gemma training implementation

#117 bzantium closed 4 months ago
0
support Qwen2 model

#116 bzantium closed 4 months ago
0
make progress bar works properly when num_proc > 1

#115 bzantium closed 4 months ago
0
show progress when num_proc > 1

#114 bzantium closed 4 months ago
0
Request for code in Appendix F

#113 Wintoplay opened 4 months ago
1
Fine Tuning Script for Token Classification

#112 WorldHellow closed 4 months ago
2
Test llama and Mistral on mteb benchmark

#111 NouamaneELGueddarii closed 4 months ago
3
Learning implications for loss_scale

#110 daegonYu closed 4 months ago
2
Fine Tuning Script on Custom Data

#109 WorldHellow closed 4 months ago
1
fix _convert_to_str function when instruction is empty

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

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

#106 daegonYu closed 4 months ago
2
Some words can not be encoded (upper case and lower case)

#105 wyhwhy closed 4 months ago
4