Comments (1)
ThePerfectComputer
commented on May 18, 2024
I noticed that running the one_file_ref which doesn't make use of xformers was actually faster on multiple GPUs than main.py which does call xformers from mistral/model.py under the hood. I too have a (slightly unrelated) architectural question about how to achieve the 131k token attention for the Mistral team which can be found in the following link. Unfortunately, it seems the Mistral team hasn't been responding to any of the Gituhb issues for this Mistral repo.
from mistral-src.
Related Issues (20)
- "evaluation pipeline" public?
- [Mistral 7B mistral-7b-instruct-v0.1.Q8_0.gguf] Wrong text "quoted" while presented as real HOT 3
- Friendly Reminder while Generating the output
- Evaluation Pipeline
- Mistral's tokenizer is not optimized
- [MISTRAL AI ERROR] Mistral AI responding with Unexpected role RoleEnum.tool error HOT 2
- CUDA EXTENSION NOT INSTALLED nvcr.io/nvidia/pytorch:22.12-py3 HOT 1
- Training code HOT 1
- Question about Mixtral MLP section
- Missing the params.json HOT 2
- Fine Tuning Mistral 7b HOT 3
- I am unable to build the vLLM Container HOT 4
- Not completing answer
- Did you run some cuda functions before calling NumCudaDevices() that might have already set an error? Error 804: forward compatibility was attempted on non supported HW HOT 1
- How to use a prompt for text analysis? HOT 6
- JSON response format failing to retrieve clean JSON
- Q: Why rotary embedding applied only to queries and keys?
- Did Mistral-7B-Instruct-v0.2 use Sliding Window Attention (SWA)?
- AI generates responses or conversation without any human input
- PAD token missing ? HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from mistral-src.