YuJuLin's picture

17 5

YuJuLin

nps798

·

nps798

AI & ML interests

None yet

Organizations

None yet

nps798's activity

New activity in K024/mt5-zh-ja-en-trimmed about 2 months ago

Brilliant !

#3 opened about 2 months ago by

New activity in ccrains/larson-gemma-2b-chinese-v0.1 4 months ago

Looking into training detail

#1 opened 4 months ago by

New activity in BAAI/bge-large-en 5 months ago

Why do you add a normalize layer in the end of the model? Does it affect how we fine tune results?

#13 opened 5 months ago by

New activity in yuuko-eth/Rain-2x7B-MoE-32k-v0.1 5 months ago

How do you build your own MoE model?

#1 opened 6 months ago by

New activity in openai/whisper-medium 6 months ago

The same audio. Why are the results on Hugging Face different from the results of the model on GitHub?

#31 opened 7 months ago by

New activity in charent/Phi2-Chinese-0.2B 7 months ago

厲害！請教預訓練所需硬體資源以及訓練代碼原理

#2 opened 7 months ago by

New activity in stabilityai/japanese-stablelm-3b-4e1t-base 7 months ago

Code to continue pretrain

#1 opened 7 months ago by

New activity in mistralai/Mistral-7B-v0.1 9 months ago

Has a massive repetition problem

#29 opened 10 months ago by

New activity in Vasanth/phi-1_5-finetuned-gsm8k 9 months ago

finetuning

#1 opened 10 months ago by

New activity in migtissera/SynthIA-7B-v1.3 10 months ago

Hi what did you train this model with, and what were hyperparams?

#1 opened 10 months ago by

New activity in teknium/Mistral-Trismegistus-7B 10 months ago

Excellent model ! Asking about training details

#3 opened 10 months ago by

New activity in mistralai/Mistral-7B-v0.1 10 months ago

QLORA fine tuning with longer length of sequence (max_length=2048, padding=True) cause RuntimeError: CUDA error: device-side assert triggered; shorten length to 512 works !

#46 opened 10 months ago by

New activity in mistralai/Mistral-7B-Instruct-v0.1 10 months ago

How is this model different from Llama 2-7B?

#8 opened 10 months ago by

New activity in yentinglin/Taiwan-LLaMa-v1.0 11 months ago

Curious about how the model was trained to support Taiwan Chinese so well

#1 opened 11 months ago by

New activity in TheBloke/Llama-2-70B-Chat-GGML 12 months ago

using oobaboga to load model fail for 70b chat ggml Q2_k anad Q3_K_S

#2 opened 12 months ago by

New activity in mosaicml/mpt-7b about 1 year ago

After installing triton, running pipe() return "fatal error: cuda.h: No such file or directory " and "CalledProcessError: Command '['/usr/bin/gcc'...."

#66 opened about 1 year ago by