philschmid/sagemaker-huggingface-llama-2-samples issues and pull requests

#22 - The model did not return a loss from the inputs

Issue - State: closed - Opened by EtienneFerrandi about 1 year ago

#21 - No matching distribution found for peft==0.4.0

Issue - State: closed - Opened by EtienneFerrandi about 1 year ago - 4 comments

#20 - ValueError: expected sequence of length 2048 at dim 1 (got 0)

Issue - State: open - Opened by vipinap98 about 1 year ago

#19 - HF token leak

Issue - State: open - Opened by tcapelle about 1 year ago - 1 comment

#18 - Does we need the prompt formatting during the instruct model fine-tuning?

Issue - State: open - Opened by mariababich about 1 year ago

#17 - Error running fine-tuning on SageMaker

Issue - State: closed - Opened by abhyudit309 about 1 year ago - 4 comments

#16 - Does the inference notebook work with Code Llama models?

Issue - State: open - Opened by jackkwok about 1 year ago - 4 comments

#15 - Fine Tuning LLAMA2-13b on ml.g4dn.12xlarge is taking too much time.

Issue - State: closed - Opened by hz-nm about 1 year ago - 5 comments

#14 - There should be a space before the </s> tag for assistant messages

Pull Request - State: closed - Opened by mukundt about 1 year ago

#13 - Error in build_llama2_prompt()

Issue - State: closed - Opened by pleonova over 1 year ago - 1 comment

#12 - pretrain llama-2

Issue - State: open - Opened by enpassanty over 1 year ago

#11 - Confused about chunk and potential concern of it hurting finetuning performance

Issue - State: open - Opened by kalvin1024 over 1 year ago

#10 - Cannot save checkpoints on provided s3 bucket

Issue - State: closed - Opened by AnnamariaGalante over 1 year ago - 4 comments

#9 - Error running trained model on sagemaker endpoint

Issue - State: open - Opened by sdkramer10 over 1 year ago - 3 comments

#8 - RuntimeError: cannot reshape tensor

Issue - State: open - Opened by mikewadhera over 1 year ago - 1 comment

#7 - trainer.evaluate() with run_clm.py

Issue - State: open - Opened by sadransh over 1 year ago

#6 - 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'

Issue - State: open - Opened by sashakttripathi over 1 year ago - 2 comments

#5 - Inference generating output with <unks> only with batch decoding but generates correct result if I pass a single prompt.

Issue - State: open - Opened by irshadbhat over 1 year ago - 2 comments

#4 - CUDA error encountered on both BS=2&3 for 7B and 13B.

Issue - State: open - Opened by alfredcs over 1 year ago - 1 comment

Ecosyste.ms: Issues

GitHub / philschmid/sagemaker-huggingface-llama-2-samples issues and pull requests