issues
search
ssbuild
/
chatglm_finetuning
chatglm 6b finetuning and alpaca finetuning
1.54k
stars
176
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
ptv2
#284
ssbuild
closed
1 year ago
0
num_layers_freeze
#283
ssbuild
closed
1 year ago
0
简化
#282
ssbuild
closed
1 year ago
0
"gradient_checkpointing": False
#281
ssbuild
closed
1 year ago
0
support accelerator trainer
#280
ssbuild
closed
1 year ago
0
support accelerator trainer
#279
ssbuild
closed
1 year ago
0
v0.2.5
#278
ssbuild
closed
1 year ago
0
v0.2.5
#277
ssbuild
closed
1 year ago
0
support ia3
#276
ssbuild
closed
1 year ago
0
0.2.4
#275
ssbuild
closed
1 year ago
0
fix slidding
#274
ssbuild
closed
1 year ago
0
update
#273
ssbuild
closed
1 year ago
0
update
#272
ssbuild
closed
1 year ago
0
deepspeed precision
#271
ssbuild
closed
1 year ago
0
fix ptv2
#270
ssbuild
closed
1 year ago
0
fix ptv2
#269
ssbuild
closed
1 year ago
0
ptv2 remove device_map
#268
ssbuild
closed
1 year ago
0
build_template
#267
ssbuild
closed
1 year ago
0
请问adalora能用deepspeed训练吗
#266
Yu-Yuqing
opened
1 year ago
0
update
#265
ssbuild
closed
1 year ago
0
LoRA和ptv2微调均发生OOM
#264
shenzhyzzz
opened
1 year ago
4
0.2.0
#263
ssbuild
closed
1 year ago
0
0.1.21
#262
ssbuild
closed
1 year ago
0
INFO:lightning_fabric.utilities.rank_zero:`Trainer.fit` stopped: No training batches.
#261
hasakikiki
opened
1 year ago
2
有谁用过Mac Studio微调的
#260
xsailor511
opened
1 year ago
0
怎么控制每训练n轮就保存一次模型呢
#259
tjulh
closed
1 year ago
1
AttributeError: module 'torch.optim' has no attribute 'adam'
#258
evanweiguohua
opened
1 year ago
5
推理时怎么指定用哪几张卡
#257
tjulh
closed
1 year ago
2
修改max_seq_length好像并没有生效？
#256
tjulh
closed
1 year ago
4
AttributeError: module 'inspect' has no attribute 'ArgSpec'
#255
SeekPoint
closed
1 year ago
1
显示可训练参数数量问题
#254
xxll88
opened
1 year ago
0
缺省Lora训练显存消耗 60G
#253
is
opened
1 year ago
0
您好，非常感谢您的工作。我进行全参数微调后，调用infer_finetuning.py的时候，出现Missing key(s) in state_dict: "_TransformerLightningModule__backbone.transformer.lm_head.weight".请问您有遇到过这个问题吗
#252
Xuan-ZW
closed
1 year ago
2
fix potential expand vocab_size
#251
ssbuild
closed
1 year ago
0
requirements.txt
#250
ssbuild
closed
1 year ago
0
load float16 weight
#249
ssbuild
closed
1 year ago
0
support resize embs
#248
ssbuild
closed
1 year ago
0
模型训练只使用到了单个GPU
#247
GZJAS
closed
1 year ago
1
0.1.10
#246
ssbuild
closed
1 year ago
0
ptuning v2 如何启动quantization_bit 4
#245
xxll88
opened
1 year ago
1
v0.1.10
#244
ssbuild
closed
1 year ago
0
用单轮数据集。 p-tuning微调chatGLM之后出现的问题。
#243
SMR-S
opened
1 year ago
1
单轮的QA回答的还不错，但是多轮对话时，就会出现瞎回答的情况。我感觉应该是多轮对话的history记录干扰了回答的效果？目前还不确定是不是这样的情况，有兄弟遇到类似的情况吗？交流一波啊！
#242
SMR-S
closed
1 year ago
0
should be load_sft_weight?
#241
HenryYuxuanWang
closed
1 year ago
1
运行lora的训练代码， int8=True ，推理时报错 RuntimeError: expected scalar type Half but found Float，这是什么原因？
#239
MathamPollard
closed
1 year ago
14
执行infer_lora_finetuning.py报错：‘NoneType’ objectg has no attribute 'learning_rate'
#238
paizhongxing
closed
1 year ago
8
input_ids格式是否需要<CLS>
#237
Jong-Won
opened
1 year ago
0
如何使用evaluate.py对测试集进行验证
#236
lawrencelxy
opened
1 year ago
4
Overriding torch_dtype=None with `torch_dtype=torch.float16` due to requirements of `bitsandbytes` to enable model loading in mixed int8. Either pass torch_dtype=torch.float16 or don't pass this argument at all to remove this warning.
#235
sanwei111
opened
1 year ago
0
关于需要多少显卡资源
#234
sanwei111
opened
1 year ago
1
Next