Eike Steffen Kohlmeyer

icon indicating a website link https://thebrights.ai

icon company Freelancer icon location Munich icon location Machine Learning Enginner with focus on NLP and LLM

Results 2 issues of Eike Steffen Kohlmeyer

Missing key(s) in state_dict for bias in attention blocks

1 comment

I am trying to run step 3 of the RLHF examples using a RewardModel checkpoint that I trained using step 2 of the examples. For every step, I used the...

Step 3: RuntimeError: CUDA error: misaligned address

3 comment

I try to run RLHF for my previously trained Actor and Reward model. However, I encounter the following Exception: ``` Traceback (most recent call last): File "/home/ec2-user/SageMaker/deepspeedexamples-fork/applications/DeepSpeed-Chat/training/step3_rlhf_finetuning/main.py", line 516, in...