Hanif Yuli Abdillah P

Exalate commented: hanifabd commented: @jakkritz could i join to your repo? i got problem in custom action. my connectors didn't want to receive messages from custom action. thank you. here...

I applied it for sentence boundary disambiguation, when i remove the bias, my model be more aggresive in segmenting the sentence but when i add bias = 1 it make...

getting the same issues when i try with wikipedia url. i use this sample link "https://id.wikipedia.org/wiki/Soekarno". have you find how to tackle this issue?

Haloo @danielhanchen , I was using this notebook provided here [Mistral Notebook](https://colab.research.google.com/drive/1_yNCks4BTD5zOnjozppphh5GzMFaMKq_?usp=sharing) i changed this `!pip install "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git"` to be like this below. ``` !pip uninstall unsloth -y...

FYI: I was successfully load `Mistral-7B-Instruct-v0.2` from mistralai repo by adding my huggingface token with this code, but get error when define trainer using SFTTrainer still get error when loading...

> Are you doing `load_in_4bit = False`? This will consume 4x more memory i tried everything include that, but still failed, is there any example to load `unsloth/mistral-7b-instruct-v0.2`?

i used the provided notebook. when i changed to `unsloth/mistral-7b-instruct-v0.2` and run in on google colab, i directly got that error and no error pop up about out of memory...