Giters
mallorbc
/
Finetune_LLMs
Repo for fine-tuning Casual LLMs
Geek Repo:
Geek Repo
Github PK Tool:
Github PK Tool
Stargazers:
449
Watchers:
10
Issues:
20
Forks:
83
mallorbc/Finetune_LLMs Issues
torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 394.00 MiB
Updated
7 months ago
Comments count
3
Unable to find image 'gpt:latest' locally
Closed
7 months ago
Comments count
1
gradient overflow when training 13b Llama Model on 7 a100s
Updated
10 months ago
Comments count
1
Can't find a valid checkpoint
Closed
10 months ago
Comments count
1
Sends Kill to process when trying to resume a finetune on LLaMA 7B
Closed
10 months ago
Comments count
2
RuntimeError: Error building extension 'cpu_adam'
Closed
10 months ago
Comments count
5
RuntimeError: The expanded size of the tensor (50257) must match the existing size (0) at non-singleton dimension 0. Target sizes: [50257]. Tensor sizes: [0]
Closed
10 months ago
Comments count
2
`RuntimeError: Error building extension 'cpu_adam'AttributeError: 'DeepSpeedCPUAdam' object has no attribute 'ds_opt_adam'
Closed
10 months ago
Comments count
1
"nvcc fatal : Unsupported gpu architechture 'compute_89'" with docker image
Closed
a year ago
Comments count
3
cannot import name 'GPTNeoXForCausalLM' from 'transformers'
Closed
a year ago
Comments count
1
Running super slow on 4 a100 gpus
Closed
a year ago
Comments count
2
How to make the inference of GPT-J run on multiple GPU ?
Closed
2 years ago
Comments count
2
DeepSpeedZeRoOffload initialize [end]
Closed
2 years ago
Comments count
4
[QUESTION] single_texts vs group_texts
Closed
2 years ago
Comments count
2
File: Dockerfile Line:32
Closed
2 years ago
Comments count
1
Incorrect block size?
Closed
2 years ago
Comments count
3
Training data format for generating Scenario based MCQ's
Closed
2 years ago
Comments count
2
Can't perform example_run, getting an error after deepspeed is initialized
Closed
3 years ago
Comments count
2
deepspeed>=0.5.7 is required by recent versions of the transformers package
Closed
3 years ago
Comments count
3
Error while running convert_model_to_torch script
Closed
3 years ago
Comments count
3