ise-uiuc / magicoder

Magicoder: Source Code Is All You Need

https://arxiv.org/abs/2312.02120

ise-uiuc/magicoder Issues

Achieved close performance of MagicoderS by finetuning only with `evol-codealpaca-v1`.
Closed 3 months ago8
Quantised Finetuning on 22GB*4 GPUs
Closed a month ago
can you consider adding my explanation on how to use magicoder in text-generation-webui
Updated a month ago1
Got same problem that model only return lots of '\n'
Updated 2 months ago5
HuggingFace Playground has failed
Closed 2 months ago
Is it normal to take more than one hour to get the humanevalplus results?
Closed 2 months ago
8台A40机器上复现magicoder-S-DS-6.7B的结果
Closed 2 months ago
The templates used in reproducing the eval results: why adding the instruction again after "### Response: "?
Updated 2 months ago
catastrophic forgetting problem
Updated 2 months ago
A scaling law of instruction-code-data would be very interesting...
Updated 3 months ago3
How to write prompt for code completion task
Closed 3 months ago1
used Dilated attenton instead of Vanilla Attention in Llama model and fine-tuen the model ,
Updated 3 months ago
`OSS-Instruct` dataset generation instructions
Closed 3 months ago1
The correctness of solution
Closed 3 months ago1
Are the training loss and validation loss recorded?
Closed 3 months ago4
Data collection and generation
Closed 3 months ago1
How do you set the 'stop_words' parameter
Updated 3 months ago
The model outputs nothing but "\n" HELP! 😭
Closed 5 months ago4
Training data format for Magicoder-OSS-Instruct-75K
Closed 5 months ago4
Confusion about the code of train
Closed 5 months ago4
Does this model support fill-in-the-middle?
Closed 5 months ago3
So many impressive experiments ! Are there any experiments with neftune ?
Closed 5 months ago1
small models
Closed 5 months ago1
any environment requirement for the model, doesn't work in MacAir M1 (16G)
Updated 5 months ago1
Magicoder-S-DS-6.7B gguf, gptq models are not working. Please could you provide quantize models. LM studio, text generation web ui is not is not working.
Updated 5 months ago1
How do you test HE after fine-tuned on CodeLLama？
Closed 5 months ago1
Text-gen prompt template?
Updated 5 months ago3
Possibility for a Mixture-of-Experts Model?
Closed 5 months ago1
Any plans for a 33b fine tune?
Closed 5 months ago2
Max Token = ?
Closed 5 months ago1
Optimizer selection
Closed 5 months ago2
Wrong MBPP pass@1 of CodeLlama-Python
Closed 5 months ago1
the evaluation result of magicoder is not aligned with the result on paper
Closed 5 months ago7
Evaluation codes not found
Closed 5 months ago3
can I use vLLM to load the models?
Closed 5 months ago1