-
From the README
"If you try an unsupported model, you'll see "gibberish output".
This happens for instance with https://huggingface.co/TheBloke/vicuna-13B-1.1-GPTQ-4bit-128g
If you know how to use …
-
### Issue you'd like to raise.
Hi,
the gpt4all-chat source code needs some minor changes to compile on Debian Bullseye (actually [Devuan Chimaera](https://www.devuan.org/os/announce/chimaera-relea…
-
### Describe the bug
I use [text-generation-webui](https://github.com/oobabooga/text-generation-webui) and use [llama.cpp](https://github.com/ggerganov/llama.cpp) as loader for the GGML format La…
-
From reading https://huggingface.co/TheBloke/wizard-vicuna-13B-SuperHOT-8K-GPTQ I get the sense that exllama supports context sizes greater than 2k - as part of oogabooga, anyhow.
When I load the m…
-
# Prerequisites
- [x] I am running the latest code. Development is very rapid so there are no tagged versions as of now.
- [x] I carefully followed the [README.md](https://github.com/ggerganov/llama…
-
**To Reproduce**
I am using the WizardCoder [training script](https://github.com/nlpxucan/WizardLM/blob/main/WizardCoder/src/train_wizardcoder.py) to further fine-tune the model on some examples th…
-
## Is your feature request related to a problem? Please describe.
- STORY: Users can generalize about dynamical systems with insufficient samples (and Anosov flows)
## Describe the solution you'd …
-
Run in nvidia pytorch docker 23.05, with rtx a6000 48G
```
from transformers import LlamaForCausalLM, LlamaTokenizer
import torch
ckpt = '/root/hdd/WizardLM-13B-V1.0'
device = torch.device('…
-
### Overview
The documentation states that it is possible to use any OpenAI-compatible API.
As I have a local working installation of `text-generation-webui` I attempted to use this with my alread…
-
### Describe the bug
1. Aitrepreneur_wizardLM-7B-GPTQ-4bit-128g
2. facebook_opt-350m
3. TheBloke_stable-vicuna-13B-GPTQ
Which one do you want to load? 1-3
3
Loading TheBloke_stable-vicun…