-
Tested on starcoder-16b. It consistently completes `if █` string incorrectly without parentheses around a condition:
```ts
if featureFlagProvider.enabled {
```
This behavior is so consistent t…
-
Hello,
I am running inferences on StarCoder on a 112GB RAM CPU cluster. While asking StarCoder to help find some issues in my code, it highlights possible errors but it also generates some junk out…
-
**What problem or use case are you trying to solve?**
Currently OpenDevin somewhat works with the strongest closed LLMs such as GPT-4 or Claude Opus, but we have not confirmed good results with ope…
-
I am exploring the possibility of using StarCoder to generate embeddings for code tokens and would like to know if this is feasible with the current implementation.
### Questions:
1. Is it possib…
-
Hi, I have a question about the WizardCoder model, especially in 1B size. I read the paper about WizardCoder and can see that WizardCoder-15B is fine-tuned from StarCoder 15B. I want to ask whether th…
-
We observed noticeable variability when re-running the FSDP model training script for a small 1.xB llama2 model with fixed seed(s) and same tokens. Below is a snapshot of the evaluation results on thr…
-
Hello,
I have been trying to use the finetune.py script with my own dataset on a single H100 GPU with CUDA 11.8
I have been getting the following error.
The same script and dataset are working wh…
-
**Is your feature request related to a problem? Please describe.**
Thank you for putting this together. It helped me a lot to learn the big picture of LLMs.
I tried to build and run it on an…
-
When I activate the local execution, I get the following error message:
`
ValueError: The current "device_map" had weights offloaded to the disk. Please provide an "offload_folder" for them. Alterna…
-
I have written a usable fill-in-the-middle function. The code is available [here](https://github.com/NightMachinery/doom.d/blob/master/autoload/night-ellama.el#L92) (look for `night/ellama-code-fill-i…