molohov

Results 16 comments of molohov

Here is the explicit Anaconda environment if you'd like to try reproducing it in Windows: ``` # This file may be used to create an environment using: # $ conda...

Interestingly this only happens with ctrl. It does not happen with alt or cmd

Here is the environment.yml that I use: ``` name: keylogger channels: - defaults dependencies: - pip - pip: - pynput ```

Did you ever figure this out? I'm also trying to figure out how to get this working with Vivaldi profiles.

Any update on automatic rules? I'd help, but I have no experience with C# :)

@vlkv could you also add the svgs for v2 and your latest v4 also?

I am seeing this error too. DeepSpeed version 0.9.2 ``` config = AutoConfig.from_pretrained(model_name) with deepspeed.OnDevice(dtype=dtype, device="meta"): model = AutoModelForCausalLM.from_config(config) model = deepspeed.init_inference( model, tensor_parallel = tp_config, base_dir=repo_root, replace_with_kernel_inject=args.kernel_injection, **kwargs )...

I had some success loading the model this way: ``` with deepspeed.OnDevice(dtype=dtype, device="meta"): model = AutoModelForCausalLM.from_pretrained(model_name, low_cpu_mem_usage=True) model = deepspeed.init_inference( model, tensor_parallel = tp_config, base_dir=repo_root, replace_with_kernel_inject=args.kernel_injection, **kwargs ) ``` I...

> I can confirm that I'm able to replicate this. Interestingly, I'm finding that smaller OPT models work loading with meta tensor. It appears that models that are split in...