philschmid sagemaker-huggingface-llama-2-samples issues

philschmid / sagemaker-huggingface-llama-2-samples

86 stars 32 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

The model did not return a loss from the inputs

#22 EtienneFerrandi closed 10 months ago
0
No matching distribution found for peft==0.4.0

#21 EtienneFerrandi closed 10 months ago
4
ValueError: expected sequence of length 2048 at dim 1 (got 0)

#20 vipinap98 opened 11 months ago
0
HF token leak

#19 tcapelle opened 11 months ago
1
Does we need the prompt formatting during the instruct model fine-tuning?

#18 mariababich opened 1 year ago
0
Error running fine-tuning on SageMaker

#17 abhyudit309 closed 1 year ago
4
Does the inference notebook work with Code Llama models?

#16 jackkwok opened 1 year ago
4
Fine Tuning LLAMA2-13b on ml.g4dn.12xlarge is taking too much time.

#15 hz-nm closed 1 year ago
5
There should be a space before the </s> tag for assistant messages

#14 mukundt closed 1 year ago
0
Error in build_llama2_prompt()

#13 pleonova closed 1 year ago
1
pretrain llama-2

#12 enpassanty opened 1 year ago
0
Confused about chunk and potential concern of it hurting finetuning performance

#11 kalvin1024 opened 1 year ago
0
Cannot save checkpoints on provided s3 bucket

#10 AnnamariaGalante closed 1 year ago
4
Error running trained model on sagemaker endpoint

#9 sdkramer10 opened 1 year ago
3
RuntimeError: cannot reshape tensor

#8 mikewadhera opened 1 year ago
1
trainer.evaluate() with run_clm.py

#7 sadransh opened 1 year ago
0
OSError: meta-llama/Llama-2-13b-hf is not a local folder and is not a valid model identifier listed on 'https://huggingface.co/models'

#6 sashakttripathi opened 1 year ago
2
Inference generating output with <unks> only with batch decoding but generates correct result if I pass a single prompt.

#5 irshadbhat opened 1 year ago
2
CUDA error encountered on both BS=2&3 for 7B and 13B.

#4 alfredcs opened 1 year ago
1
RuntimeError: mat1 and mat2 shapes cannot be multiplied (4096x5120 and 1x2560)

#3 monuminu opened 1 year ago
1
No models to quantize

#2 andreaolgiati closed 1 year ago
2
missing "requirements.txt"

#1 andreaolgiati closed 1 year ago
5