metadata
license: apache-2.0
language:
- en
tags:
- merge
Test merge. Attempt to get good at RP, ERP, general things model with 128k context. Every model here has Epiculous/Fett-uccine-Long-Noodle-7B-120k-Context in merge instead of regular MistralYarn 128k. The reason is because i belive Epiculous merged it with Mistral Instruct v0.2 to make first 32k context as good as possible, if not than it's sad.
Here is the "family tree" of this model, im not writing full model names cause they long af
NeuralKunoichi-EroSumika 4x7B
* NeuralKunoichi-EroSumika 4x7B
*(1) Kunocchini-7b-128k
|
*(2) Mistral-Instruct-v0.2-128k
* Mistral-7B-Instruct-v0.2
|
* Fett-128k
|
*(3) Erosumika-128k
* FErosumika 7B
|
* FFett-128k
|
*(4) Mistral-NeuralHuman-128k
* Fett-128k
|
* Mistral-NeuralHuman
* Mistral_MoreHuman
|
* Mistral-Neural-Story