microsoft
/

Phi-3-mini-4k-instruct

Text Generation

Inference Endpoints

Model card Files Files and versions Community

Resources

View closed (52)

Jetson nano

#74 opened about 19 hours ago by

fixed generation_args in Sample inference code

#73 opened 7 days ago by

Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.

#72 opened 9 days ago by

tflite convertion

#71 opened 13 days ago by

why EOS=<|endoftext|> not </s> when <s> is used as BOS?

#70 opened 17 days ago by

fine-tuning with structured data set

#68 opened 22 days ago by

Questions about chat template format (whether to escape newline characters and special characters or not)

#67 opened 25 days ago by

Datasets used for training

#63 opened about 1 month ago by

saransh03sharma

fp16 normal weights

#62 opened about 1 month ago by

GREAT MODEL

#61 opened about 1 month ago by

Sliding window = 2047?

#60 opened about 1 month ago by

Recent change on the rstrip property on special tokens

#59 opened about 1 month ago by

fix: update tokenizer config to support `add_generation_prompt=True` and clarify content

#57 opened about 1 month ago by

Fix auto-casting with SFT sometimes only up-casting keys and not queries

#56 opened about 1 month ago by

Help with merging LoRA layers back onto Phi3

#55 opened about 1 month ago by

Changed instruction/chat template

#54 opened about 1 month ago by

Phi-3 support for GPU training with MPS on Mac

#52 opened about 2 months ago by

System prompts ignored in chat completions

#51 opened about 2 months ago by

Fine-tuning is not improving the domain knowledge? it is very complicated, could you help?

#50 opened about 2 months ago by

the phi3 model not learning from training data using orpo.

#44 opened about 2 months ago by

Leaking Training Data

#40 opened about 2 months ago by

Instruct model needs work

#35 opened about 2 months ago by