nguyenbh (Nguyen Bach) – Community Activity

New activity in microsoft/Phi-3.5-MoE-instruct about 1 month ago

#2 opened 3 months ago by

#1 opened 3 months ago by

#43 opened about 2 months ago by

#91 opened 3 months ago by

#90 opened 3 months ago by

#83 opened 5 months ago by

#82 opened 5 months ago by

#81 opened 5 months ago by

#80 opened 5 months ago by

#79 opened 5 months ago by

#77 opened 5 months ago by

#75 opened 5 months ago by

#74 opened 5 months ago by

#72 opened 5 months ago by

#66 opened 6 months ago by

New activity in microsoft/Phi-3.5-MoE-instruct about 1 month ago

#46 opened about 1 month ago by

New activity in microsoft/Phi-3.5-mini-instruct about 1 month ago

#22 opened about 2 months ago by

#5 opened 3 months ago by

New activity in microsoft/Phi-3-mini-4k-instruct about 2 months ago

#90 opened 4 months ago by

#96 opened 2 months ago by

Nguyen Bach

nguyenbh's activity

Nguyen Bach

AI & ML interests

Organizations

nguyenbh's activity

About training mem usage for moe

CPU support by ONNX as phi3?

VRAM requirements

deployment issue

Update sample_finetune.py

I'm running out of memory while generating on a RTX A5000 (24 GB)

sample code throws TypeError: can only concatenate str (not "dict") to str

"TypeError: Phi3 is not supported yet." while doing awq quantization using autoAWQ

Train a BitNet version of Phi 3 Mini?

Not implemented the non-interpolated starting token?

raise FileMetadataError("Distant resource does not have a Content-Length.") huggingface_hub.utils._errors.FileMetadataError: Distant resource does not have a Content-Length.

Update the tokenizer which inserted added_tokens to vocab

model.generate() is crashing

I am trying to use this for some coding tasks (in particular for Java language) and I am getting borderline nonsensical results. Has someone else faced this issue?

NER using phi3

system messages

Discrepency in template

`TypeError` When Processing Text and Image Batch with `processor`

add tgi tag

Update README.md