mamba
mamba copied to clipboard
state-spaces
Hello, if I don’t have the permissions to update GLIBC under the root directory, how can I solve this problem?
i browse mamba block code and have some questions: 1. https://github.com/state-spaces/mamba/blob/12d855003ba92c8a15d1739ce65a14c6fb16e254/mamba_ssm/modules/mamba_simple.py#L92C5-L99C44 https://github.com/state-spaces/mamba/blob/12d855003ba92c8a15d1739ce65a14c6fb16e254/mamba_ssm/modules/mamba_simple.py#L240 in line 92-99, copy $softplus\ inverse(dt\_ bias)$ to dt_proj.bias, dt_bias~uniform(d_min, d_max) . and in line 240, $dt =...
Dear @tridao , @albertfgu , It looks like it is not straightforward to generate onnx file due to following reason using torch.onnx.export: 1) It looks like the underlying scan operator...
Hi, author, I would like to install selective_scan_cuda on a Windows system, how should I install it, is there currently only one way to install it on Linux?
I'm not sure which community I need to turn to for help, so I'm posting them all, I would appreciate it if someone could answer my questions: [LoRA layer param.grad=None...
I'm looking to accomplish some sort of Vector to Vector task with Mamba. Does any encoder-decoder architecture exist, or any alternative approaches to this task using Mamba? Or is having...
Maybe there are some benifit below: 1, The code could be simplier. 2, The inference could be faster. 3, The inference can accept multi-tokens in this way. There are some...
Hello I was wondering about params.n_chunks. What does the n_chunks variable do, and why does it have anything to do with the length of the sequence
When running `python evals/lm_harness_eval.py --model mamba --model_args pretrained=state-spaces/mamba-130m --tasks lambada_openai,hellaswag,piqa,arc_easy,arc_challenge,winogrande --device cuda --batch_size 64` After a little bit, I get : ``` 2024-03-22:11:33:09,519 INFO [evaluator.py:362] Running loglikelihood requests Traceback (most...
