rain1

difference in compiled output for star.tbf1 program

2

Hello! I created some terminology: * t1: running a tbf1 script in the python prototype * t2: compiling a tbf1 script with the tinyboot1.tbf1 script running on top of the...

error I got running instructions

1

``` Gems updated: benchmark bigdecimal bundler cgi csv date delegate did_you_mean etc fcntl fiddle fileutils forwardable getoptlong io-console ipaddr irb reline json logger matrix mutex_m digest net-pop net-protocol net-smtp observer...

Update qlora.py

1

add n_gpus = 0 default value to get_accelerate_model Fixes this error: Traceback (most recent call last): File "qlora.py", line 841, in train() File "qlora.py", line 704, in train model, tokenizer...

Error in the guanaco-7B-demo-colab.ipynb colab

1

https://colab.research.google.com/drive/17XEqL1JcmVWjHkT-WczdYkJlNINacwG7 RuntimeError: CUDA error: CUBLAS_STATUS_NOT_INITIALIZED when calling `cublasCreate(handle)` 

tail call optimization?

1

is there any intention to support proper tail calls? (iterative loops)