Giters
THUDM
/
GLM
GLM (General Language Model)
Geek Repo:
Geek Repo
Github PK Tool:
Github PK Tool
Stargazers:
3047
Watchers:
46
Issues:
188
Forks:
316
THUDM/GLM Issues
Accelerate support for GLM
Updated
a year ago
将GLM-10B-chinese模型切分为MP_SIZE=8, 然后finetune seq2seq任务时,在eval阶段报错IndexError。怀疑eval没有以MP_SIZE=8方式运行
Updated
9 months ago
Comments count
7
你好,我在使用glm-10-chinese对自己数据集进行微调的时候,卡在了第1000个iteration不动了
Updated
a year ago
Comments count
5
使用GLM-10B-Chinese模型跑seq2seq的finetune脚本报错word_embeddings.weight维度不对
Closed
a year ago
Comments count
1
请问单机8卡v100 32G能跑seq2seq的fine tune吗?我跑着会work = _default_pg.barrier()
Closed
a year ago
使用p-tuning去finetune glm-large-chinese模型时 --continuous-prompt
Closed
a year ago
Comments count
3
RuntimeError: expand(torch.HalfTensor{[1025, 4096]}, size=[1]): the number of sizes provided (1) must be greater or equal to the number of dimensions in the tensor (2)
Closed
a year ago
Comments count
2
如何通过huggingface加载的模型拿到last_hidden_states?
Closed
a year ago
accelerate 找不到模型
Updated
a year ago
Comments count
7
GLM-10B 模型效率问题
Updated
a year ago
BUG: GLM-10B-Chinese model generate " ⁇".
Updated
a year ago
Comments count
3
基于Prompt数据集如何微调模型?
Updated
a year ago
How to set hyperparameters during pretraining glm_doc?
Updated
a year ago
模型并行训练结束后,如何将多个模型文件合并成一个?
Closed
a year ago
Comments count
2
基于10B模型继续预训练,遇到world size 不一致导致报错
Updated
a year ago
Comments count
2
cmrc数据集结果,预测结果都为空
Updated
a year ago
chatglm-6b
Updated
a year ago
Comments count
1
单卡pretrain chinese-large模型
Closed
a year ago
impelement by megengine
Updated
a year ago
hugging face仓库的10b-chinese模型问题。用Trainer API进行数据并行微调会报出OOM错误 ,有没有优化内存的方法?
Updated
a year ago
Comments count
3
GLM 10B 模型零样本结果无法对齐
Updated
a year ago
The attention mask and the pad token id were not set问题
Updated
a year ago
Comments count
1
Does this model support temperature and repetition_penalty?
Updated
a year ago
GPT2Dataset和BlockDataset
Updated
a year ago
小数据finetune large-chinese rouge 为0
Updated
a year ago
Comments count
2
160G内存,两张24G3090,800G硬盘的环境下,对GLM-10-chinese进行finetune
Updated
a year ago
Comments count
3
50035 token id 报错
Updated
a year ago
Comments count
1
如何使用onnxruntime 支持glm优化
Updated
a year ago
GLMForSequenceClassification的使用
Updated
a year ago
Comments count
2
AutoModelForCausalLM
Updated
a year ago
Comments count
1
bash evaluate_lm.sh问题
Updated
a year ago
GPU_id设置
Updated
a year ago
generate_samples.py问题
Updated
a year ago
模型微调
Closed
a year ago
模型微调
Closed
a year ago
Comments count
1
在测试lambada时报错,显示参数错误,分布式错误
Closed
a year ago
利用huggingface glm-10b-chinese模型,跑huggingface上面的例子
Updated
a year ago
Comments count
2
run ds_pretrain_nvidia.sh
Closed
a year ago
Comments count
2
怎么使用batch beam search
Closed
a year ago
Comments count
4
genrate_sample.py的问题
Closed
a year ago
Comments count
1
`glm-10b-chinese` `build_inputs_for_generation` missing `targets` argument
Closed
a year ago
Comments count
1
finetuing MP_SIZE问题
Closed
a year ago
Comments count
1
scripts目录下是缺少一个ds_config.json吗
Closed
a year ago
The attention_mask dimension not right?
Closed
a year ago
Comments count
1
dockerfile中的ssh-env-config.sh文件怎么配置
Closed
a year ago
Comments count
2
customization dataset在fine-tune和inference的输入不同。
Closed
a year ago
Comments count
1
请问GLM-10B-Chinese的tokenizer是否支持添加自定义的token?
Updated
a year ago
运行 ds_finetune_superglue.sh key error "dev-0"
Closed
a year ago
Comments count
1
多机并行可以给点示例吗?
Closed
a year ago
Comments count
3
请问GLM模型是否可以生成长句子?我对模型进行推理或者微调的时候都会报出维度不匹配的错误
Closed
a year ago
Comments count
3
Previous
Next