Update README.md

README.md

@@ -30,17 +30,21 @@ Next projects will involve that model as the "smarts pillar" of further merges,
 # further developpements
 
 The adventure continues with Doberman_V1, a Hermes flavored Dobby on Smarteaz abliterated steroids (very good at being "in character") :
-- Nexesenex/Llama_3.x_70b_Doberman_V1 : https://huggingface.co/Nexesenex/Llama_3.x_70b_Doberman_V1 (less than 3.40 ppl 512 wiki-eng)
+- Nexesenex/Llama_3.x_70b_Doberman_V1 : https://huggingface.co/Nexesenex/Llama_3.x_70b_Doberman_V1 (less than 3.40 ppl 512 wiki-eng, -0.07 compared to Smarteaz_V1)
 
 And the saga continues again with 
 
-Nemesis_V1, a Hermes flavored Negative Llama on Smarteaz abliterated steroids (More stiff and less creative than Doberman, though) :
-- Nexesenex/Llama_3.x_70b_Nemesis_V1 : https://huggingface.co/Nexesenex/Llama_3.x_70b_Nemesis_V1 (less than.. 3.35 ppl 512 wiki-eng)
+Nemesis_V1 (ex Negames), a Hermes flavored Negative Llama on Smarteaz abliterated steroids (More stiff and less creative than Doberman, though) :
+- Nexesenex/Llama_3.x_70b_Nemesis_V1 : https://huggingface.co/Nexesenex/Llama_3.x_70b_Nemesis_V1 (less than.. 3.35 ppl 512 wiki-eng, -0.05 compared to Doberman_V1)
 
-Evasion_V1, a Hermes flavored Eva_01 on Smarteaz abliterated steroids (in test) :
-- Nexesenex/Llama_3.x_70b_Evasion_V1 : https://huggingface.co/Nexesenex/Llama_3.x_70b_Evasion_V1 (less than 3.40 ppl 512 wiki-eng, -0.02 compared to Doberman)
+Evasion_V1 (ex Hermeva), a Hermes flavored Eva_01 on Smarteaz abliterated steroids (the most creative) :
+- Nexesenex/Llama_3.x_70b_Evasion_V1 : https://huggingface.co/Nexesenex/Llama_3.x_70b_Evasion_V1 (less than 3.40 ppl 512 wiki-eng, -0.02 compared to Doberman_V1)
 
-I think that any model can be tweaked the same way, gaining "intelligence" and "quality" at the cost of a bit of its initial personality.
+Trinity_V1, a merge of Evasion as base, Doberman and NegaTessTease to include a bit of Tess (to be tested) :
+- Nexesenex/Llama_3.x_70b_Trinity_V1 : https://huggingface.co/Nexesenex/Llama_3.x_70b_Trinity_V1 (less than 3.40 ppl 512 wiki-eng, -0.03 compared to Doberman_V1)
+
+I think that most models can be tweaked the same way, with triple stock merges interlacing intruct finetunes and base finetunes.
+- This, gaining overall "intelligence" and "quality" at the cost of a bit of its initial instructions, flavor and "personality".
 
 ---
 # credits