-
As far as I know the BLOCKLIST finish reason is a fairly new feature and I've been experiencing a lot of stopped responses with this reason lately.
Unfortunately I do not have specific steps to rep…
-
If run this code:
```python
model = Qwen2VLForConditionalGeneration.from_pretrained(
"Qwen/Qwen2-VL-7B-Instruct",
torch_dtype=torch.bfloat16,
attn_implementation="flash_attention_2",
…
-
Hi Quentin
I haven't managed to find any information about how IGoR handles allelic variants presented in models. In default models some IGHV, TRAV and TRBV genes have several alleles (up to 7) and…
-
I see that it is possible get standard errors for covariates in MHMM models.
I am wondering if it is possible to get standard errors for the transition probabilities and emission probabilities in s…
-
This is an update to issue #893 which was closed because XLA compile was failing making apples to apples comparisons impossible. XLA compile was fixed with #908 and I include updated benchmarks here.…
-
Anaconda has a license that makes using `defaults` channel problematic for organizations above 200 people (). I will make a PR removing the guidance to use the `defaults` channel. This will mean we ca…
-
### Description:
This project updates the packaging to support generating synthetic OCR training data for Tibetan script in both Tibetan Pecha and Modern Tibetan Book formats. Enhancements include up…
-
I noticed that if scores were calculated with FTR or FTN
they tend to be between 0 and 1. Well actually I also saw 1.000041.
The scores by IndicLID-BERT however are raw logits, so could for exampl…
-
Hello,
After completing HP optimisation, I extract the best_estimator_ which looks like this:
![image](https://user-images.githubusercontent.com/82275999/227367264-5dadfe24-1062-4f6f-9120-2c65b8b29e…
-
Expose the probabilities of the top tokens in a prediction result
TODO:
- [x] OpenAI
- [ ] HF Models