mamba
mamba copied to clipboard
state-spaces
@tridao Hello! I am currently working with the selective scan forward kernel, specifically the step h_t = A*h_{t-1} + Bx, where h_0 is currently set to 0. I would like...
Is it possible to utilize the Mamba model for 3D model reconstruction tasks using my own private dataset of point cloud data? If so, How can I utilize the Mamba...
when I do evaluation for lm_harness_eval, there are some error as below. what is problem in it? and how to deal with it to make it work successfully? Thanks so...
Hi, I have encountered an issue when trying to install your package with poetry as a package manager, or when trying to install with pip while using the build system...
This work is very inspiring. I have a specific questions about induction head task. 1. Can you please provide more details about the **model and experiments configuration** that you used...
I used the provided instance code and encountered this issue. How can I solve it？ “ import torch from mamba_ssm import Mamba batch, length, dim = 2, 64, 16 x...
We attempted to use the recurrent mode of Mamba model and got this bug. Directly using the forward() function is good. But if the code goes to the step() function,...
This WIP PR provides FSDP training using pytorch lightning. The current model setup provides a byte level model. I'm not sure it makes sense to merge in this state, but...
Hi，When I run Mamba in a Docker container, GPU 0 is running YOLOv8, while GPU 1 is running Mamba. However, the GPU utilization of GPU 1 is 0, while GPU...
Hello, I'm currently working with the `transformers` library to train a model on causal language modeling tasks using the `MambaForCausalLM` class. However, I've noticed that the typical approach to training...
