Nicholai/jan
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: rephrase merge process

Browse Source
This commit is contained in:
hahuyhoang411 2024-03-01 15:28:43 +07:00
parent 218866b0a7
commit 5a4f60868d
1 changed files with 5 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
docs/blog/rag-is-not-enough.md
View File

@ -31,9 +31,7 @@ Problems still arise with catastrophic forgetting in general tasks, commonly obs
## Selecting a strong foundation model ## Selecting a strong foundation model
[Mistral 7B](https://huggingface.co/mistralai/Mistral-7B-v0.1) continues to outshine [Meta's Llama-2 7B](https://huggingface.co/meta-llama/Llama-2-7b) and [Google's Gemma 7B](https://huggingface.co/google/gemma-7b) on meaningful benchmarks, so we selected this as a starting point. [Mistral 7B](https://huggingface.co/mistralai/Mistral-7B-v0.1) outshines both [Meta's Llama-2 7B](https://huggingface.co/meta-llama/Llama-2-7b) and [Google's Gemma 7B](https://huggingface.co/google/gemma-7b) in key benchmarks, making it our choice for a base model. Starting with a strong foundation like Mistral allowed us to achieve greater accuracy in our specialized adaptations.
Having a robust base model is critical. In our experiments, using Mistral as a starting point ensured the highest accuracy for subsequent specialized adaptations.
![Mistral vs LLama vs Gemma](img/mistral-comparasion.png) ![Mistral vs LLama vs Gemma](img/mistral-comparasion.png)
@ -49,13 +47,13 @@ Mistral alone has known, poor math capabilities, which we needed for our highly
*Figure 2: The merged model, Stealth, doubles the mathematical capabilities of its foundational model while retaining the performance in other tasks.* *Figure 2: The merged model, Stealth, doubles the mathematical capabilities of its foundational model while retaining the performance in other tasks.*
We found model merging to be a viable approach where each iteration is cost-effective + fast to deploy. We found merging models is quick and cost-effective, enabling fast adjustments based on the result of each iteration.
We ended up with [Stealth 7B v1.1](https://huggingface.co/jan-hq/stealth-v1.1), a [SLERP](https://github.com/Digitous/LLM-SLERP-Merge) merge of Mistral with the following: We ended up with [Stealth 7B v1.1](https://huggingface.co/jan-hq/stealth-v1.1), a [SLERP](https://github.com/Digitous/LLM-SLERP-Merge) merge of Mistral with the following:
- [WizardMath](https://huggingface.co/WizardLM/WizardMath-7B-V1.1) for its math capabilities - [WizardMath](https://huggingface.co/WizardLM/WizardMath-7B-V1.1) for its math capabilities.
- [WizardCoder](https://huggingface.co/WizardLM/WizardCoder-Python-7B-V1.0) for its coding capabilities - [WizardCoder](https://huggingface.co/WizardLM/WizardCoder-Python-7B-V1.0) for its coding capabilities.
- Our own [Trinity](https://huggingface.co/jan-hq/trinity-v1.2) model for its versatility across general tasks - Our own [Trinity](https://huggingface.co/jan-hq/trinity-v1.2) model for its versatility across general tasks.
This particular combination yielded the best tradeoff across mathematical & technical reasoning while retaining the most pre-merge performance on general tasks. This particular combination yielded the best tradeoff across mathematical & technical reasoning while retaining the most pre-merge performance on general tasks.