Issues
- 1
Inference on Custom Use Case w/o fine tuning
#23 opened by Mrs-Hudson - 0
- 3
- 2
- 4
few_shot和sft的参数选择
#24 opened by zzkzzkjsw - 1
结果并不能达到理想的回复
#18 opened by gabrielpondc - 3
- 1
关于sql generate model 训练的细节
#15 opened by dshwei - 1
schema linker 模型微调不能复现结果
#22 opened by dshwei - 1
- 1
Regarding schema filtering
#19 opened by Hari-Dorbala - 1
双向数据增强技术请教
#16 opened by mojianpo - 8
codes-7b-bird 测试结果
#14 opened by dshwei - 1
codes-7b-bird 模型在本地测试bird dev 结果复现
#13 opened by dshwei - 3
when I tokenize three datasets , There are 14029574132 tokens in the pre train corpus ,toal about 14T
#11 opened by dshwei - 3
Is it possible to upload basic models to Alibaba’s modelscope platform to facilitate downloading by domestic users?
#9 opened by CycloneBoy - 1
The size of incremental pretraining datasets.
#7 opened by binz98 - 8
模型微调
#8 opened by shellhuang1227 - 1
I have found that using the word segmentation method provided in your code, the token_id appears beyond the length of the dictionary
#10 opened by dshwei - 1
- 2
Tokenize error
#6 opened by Kaimary - 2
codes-7b-merged load error
#5 opened by ddingwang12 - 3
超出上下文限制的情况如何处理
#2 opened by steph730 - 6
- 1
Spider results
#4 opened by lwmlyy - 4
请问这里使用simCSE的目的是什么呢
#1 opened by steph730