Comments (2)
Yes, it is possible - people have had some pretty cool successes with this sort of model stacking lately.
For example, here's the config Undi95 used to make Mistral-11B-OpenOrcaPlatypus (one of the steps in his Mistral-11B-OmniMix):
slices:
- sources:
- model: Open-Orca/Mistral-7B-OpenOrca
layer_range: [0, 24]
- sources:
- model: akjindal53244/Mistral-7B-v0.1-Open-Platypus
layer_range: [8, 32]
merge_method: passthrough
dtype: bfloat16
from mergekit.
Great, thank you!
from mergekit.
Related Issues (20)
- About mergekit behavior by tokenizer HOT 7
- Adding an expert to a pre-trained Mixtral model HOT 1
- Mergekit support for GPT2 failing HOT 2
- Merging different architectures HOT 3
- RuntimeError: "addmm_impl_cpu_" not implemented for 'Half' HOT 1
- Update to accelerate==0.27.2
- Gemma support isnt working on latest verison HOT 1
- Can we add support for Whisper HOT 2
- Merges Gemma models are not loading in llamacpp-python or LM studio with Gemma support HOT 3
- Installs NVIDIA depencies on AMD GPU system wasting disk space HOT 1
- Merging quantized model with pass through HOT 2
- Have no tokenizer.model when using tokenizer_source = union HOT 2
- Unsupported architecture, GPTBigCodeForCausalLM HOT 1
- StarcoderV2 models not merging HOT 5
- Model download issue HOT 1
- OSError: bilalRahib/Tiny-LLama-NSFW-Chatbot does not appear to have a file named config.json. HOT 4
- Idea: Downscaling the K and/or Q matrices for repeated layers in franken-merges? HOT 62
- 'mergekit-yaml' is not recognized as an internal or external command HOT 1
- Fast tokenizer bad merging
- DARE TIES merge of mixed fp16 and bf16 models cannot be exl2 quanted HOT 7
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 mergekit.