-
I use WizardLM-7B-uncencored-GPTQ, pygmalion-7b-4bit-128g-cuda, pygmalion-13b-4bit-128g and PygmalionCoT-7b. All are based on LLaMA and with all I have the same problem:
All 3-5 generation there is…
-
please check this not book on kaggle getting below error
https://www.kaggle.com/code/gptforall/chatopenllm-test
from ChatOpenLLM import Chat_Llama, ChatGPTQ, Chat_AutoModels, create_tagging_cha…
-
Sorry, I am new for it.
According to the code in [inference_wizardcoder.py](https://github.com/nlpxucan/WizardLM/blob/main/WizardCoder/src/inference_wizardcoder.py), i have created a service and perf…
-
I did dare experiments on mergekit before, then I found your paper also says to experiment with mergekit, can you teach me how to run it?
The example.yml in dare is written as follows, how do I c…
-
Hi, Does it comes under GPL or MIT license.
Thanks,
Deepak
-
I have thoroughly tested MoA (with one layer) on some objective benchmarks (less subjective compared to MT-bench), such as GSM8K, HotpotQA.
It seems that when the LLMs are 7B-level, it does not work …
-
`Tool llms` require a `SchemaConverter` so that our tools JSON object aligns with their JSON schema.
Create test files `tests/llms/ShuttleAIToolModel_test` and `tests/ShuttleAISchemaConverter_test`…
-
### Describe the bug
I am not able to get any reasonable input when using codellama hosted in togetherAI.
### Reproduce
```python
import os
os.environ["TOGETHERAI_API_KEY"] = "..."
from interpre…
-
# Trending repositories for C#
1. [**BeyondDimension / SteamTools**](https://github.com/BeyondDimension/SteamTools)
__🛠「Watt Toolkit」是一个开源跨平台的多功能 Steam 工具箱。__
11 stars to…
-
Iterate on the SFT-8 dataset mixes to create pretraining and final SFT mixes for SFT-9. This requires investigating the quality and usefulness of the datasets. Community input welcome below. See the `…