Update README.md

README.md

@@ -23,7 +23,7 @@ Colab: [link](https://colab.research.google.com/drive/1qxgIPymzW6_H6s_wwXu3lknkk
 
 I've changed the prompt format from ChatML to **the original Llama-3 format in v4**. Don't forget to switch formats!
 
-**v4+**: LLama-3 prompt format:
+**v4, v5, v6+**: LLama-3 prompt format:
 ```
 <|begin_of_text|><|start_header_id|>system<|end_header_id|>
 
@@ -118,6 +118,16 @@ Saiga: В небольшой деревушке на окраине города
 
 
 ## Versions
+v6:
+- [b662833f247ca04f1843b356e7ff3ee4aef8086a](https://huggingface.co/IlyaGusev/saiga_llama3_8b/commit/b662833f247ca04f1843b356e7ff3ee4aef8086a)
+- Other name: saiga_llama3_8b_sft_m10_d1_kto_m2_d2
+- SFT dataset config: [sft_d1.json](https://github.com/IlyaGusev/saiga/blob/main/configs/datasets/sft_d1.json)
+- SFT model config: [saiga_llama3_8b_sft_m10.json](https://github.com/IlyaGusev/saiga/blob/main/configs/models/saiga_llama3_8b_sft_m10.json)
+- KTO dataset config: [pref_d2.json](https://github.com/IlyaGusev/saiga/blob/main/configs/datasets/pref_d2.json)
+- KTO model config: [saiga_llama3_8b_kto_m2.json](https://github.com/IlyaGusev/saiga/blob/main/configs/models/saiga_llama3_8b_kto_m2.json)
+- SFT wandb: [link](https://wandb.ai/ilyagusev/rulm_self_instruct/runs/0iepauzu)
+- KTO wandb: [linl](https://wandb.ai/ilyagusev/rulm_self_instruct/runs/s6l98eot)
+
 v5:
 - [d947b00c56683cd4b2f7ce707edef89318027be4](https://huggingface.co/IlyaGusev/saiga_llama3_8b/commit/d947b00c56683cd4b2f7ce707edef89318027be4)
 - KTO-tune over v4, dataset: [lmsys_clean_ru_preferences](https://huggingface.co/datasets/IlyaGusev/lmsys_clean_ru_preferences)