songkq's picture

7 6

songkq

songkq

·

AI & ML interests

None yet

Organizations

songkq's activity

New activity in 01-ai/Yi-1.5-34B-Chat-16K 6 months ago

I'm confusing if this Chat model use the standard CHATML template? Is the bos_token <|im_start|> or <|startoftext|>? Is the eos_token <|im_end|> or <|endoftext|>?

#2 opened 6 months ago by

New activity in HuggingFaceH4/zephyr-orpo-141b-A35b-v0.1 7 months ago

The model generated some unexpected results in the answer, e.g., "<|user|> 哪个步骤在最后？"

#12 opened 7 months ago by

New activity in Azure99/blossom-orca-v3 8 months ago

请问数据集的指令翻译是用的什么翻译接口？

#1 opened 8 months ago by

New activity in yuyijiong/Qwen-7b-chat-yarn-32k 10 months ago

Qwen-7b训练细节

#1 opened 10 months ago by

New activity in vonjack/Qwen-LLaMAfied-HFTok-7B-Chat over 1 year ago

How to merge the bias item of c_attn = nn.Linear(config.hidden_size, 3 * self.projection_size) into LLama?

#2 opened over 1 year ago by

New activity in Qwen/Qwen-7B-Chat over 1 year ago

Unofficial LLaMAfied Version in HF format - 非官方的LLaMA化HF格式版本

#13 opened over 1 year ago by

New activity in nvidia/nemo-megatron-mt5-3B over 1 year ago

convert nemo-megatron-mt5-3B to a binary file for triton-with-fastertransformer successfully, but tritonserver fails with undesired tensor shape

#1 opened over 1 year ago by

convert nemo-megatron-mt5-3B to a binary file for triton-with-fastertransformer successfully, but tritonserver fails with undesired tensor shape

#1 opened over 1 year ago by