--- license: apache-2.0 datasets: - Epiculous/SynthRP-Gens-v1.1-Filtered-n-Cleaned - anthracite-org/stheno-filtered-v1.1 - PJMixers/hieunguyenminh_roleplay-deduped-ShareGPT - Gryphe/Sonnet3.5-Charcard-Roleplay - Epiculous/Synthstruct-Gens-v1.1-Filtered-n-Cleaned - anthracite-org/kalo-opus-instruct-22k-no-refusal - anthracite-org/nopm_claude_writing_fixed - anthracite-org/kalo_opus_misc_240827 language: - en - fr - de - es - it - pt - ru - zh - ja pipeline_tag: text-generation tags: - merge --- ### exl2 quant (measurement.json in main branch) --- ### check revisions for quants --- ![image/png](https://cdn-uploads.huggingface.co/production/uploads/64adfd277b5ff762771e4571/P962FQhRG4I8nbU_DJolY.png) Now for something a bit different, Violet_Twilight-v0.2! This model is a SLERP merge of Azure_Dusk-v0.2 and Crimson_Dawn-v0.2! # Quants! [full](https://huggingface.co/Epiculous/Violet_Twilight-v0.2) / exl2 / [gguf](https://huggingface.co/Epiculous/Violet_Twilight-v0.2-GGUF) ## Prompting The v0.2 models are trained on ChatML, the prompting structure goes a little something like this: ``` <|im_start|>user Hi there!<|im_end|> <|im_start|>assistant Nice to meet you!<|im_end|> <|im_start|>user Can I ask a question?<|im_end|> <|im_start|>assistant ``` ### Context and Instruct The v0.2 models are trained on ChatML, please use that Context and Instruct template. ### Current Top Sampler Settings [Smooth Creativity](https://files.catbox.moe/0ihfir.json): Credit to Juelsman for researching this one!
[Variant Chimera](https://files.catbox.moe/h7vd45.json): Credit to Numbra!
[Spicy_Temp](https://files.catbox.moe/9npj0z.json)
[Violet_Twilight-Nitral-Special](https://files.catbox.moe/ot54u3.json)
## Merging The following config was used to merge Azure Dusk and Crimson Dawn ```yaml slices: - sources: - model: Epiculous/Azure_Dusk-v0.2 layer_range: [0, 40] - model: Epiculous/Crimson_Dawn-V0.2 layer_range: [0, 40] merge_method: slerp base_model: Epiculous/Azure_Dusk-v0.2 parameters: t: - filter: self_attn value: [0, 0.5, 0.3, 0.7, 1] - filter: mlp value: [1, 0.5, 0.7, 0.3, 0] - value: 0.5 # fallback for rest of tensors dtype: bfloat16 ```