**tl;dr: AlphaMonarch-7B is a new DPO merge that retains all the reasoning abilities of the very best merges and significantly improves its conversational abilities. Kind of the best of both worlds in a 7B model. 🎉** AlphaMonarch-7B is a DPO fine-tuned of mlabonne/NeuralMonarch-7B using the argilla/OpenHermes2.5-dpo-binarized-alpha preference dataset. It is based on a merge of the following models using LazyMergekit: * mlabonne/OmniTruthyBeagle-7B-v0 * mlabonne/NeuBeagle-7B * mlabonne/NeuralOmniBeagle-7B
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
Beyonder-4x7B-v3 is an improvement over the popular Beyonder-4x7B-v2. It's a Mixture of Experts (MoE) made with the following models using LazyMergekit: * mlabonne/AlphaMonarch-7B * beowolx/CodeNinja-1.0-OpenChat-7B * SanjiWatsuki/Kunoichi-DPO-v2-7B * mlabonne/NeuralDaredevil-7B Special thanks to beowolx for making the best Mistral-based code model and to SanjiWatsuki for creating one of the very best RP models. This model uses a context window of 8k. I recommend using it with the Mistral Instruct chat template (works perfectly with LM Studio).
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
输入 文本→ 输出 文本
ChimeraLlama-3-8B-v2 is a merge of the following models using LazyMergekit: * NousResearch/Meta-Llama-3-8B-Instruct * mlabonne/OrpoLlama-3-8B * cognitivecomputations/dolphin-2.9-llama3-8b * Locutusque/llama-3-neural-chat-v1-8b * cloudyu/Meta-Llama-3-8B-Instruct-DPO * vicgalle/Configurable-Llama-3-8B-v0.3
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
ChimeraLlama-3-8B-v3 is a merge of the following models using LazyMergekit: * NousResearch/Meta-Llama-3-8B-Instruct * mlabonne/OrpoLlama-3-8B * cognitivecomputations/dolphin-2.9-llama3-8b * Danielbrdz/Barcenas-Llama3-8b-ORPO * VAGOsolutions/Llama-3-SauerkrautLM-8b-Instruct * vicgalle/Configurable-Llama-3-8B-v0.3 * MaziyarPanahi/Llama-3-8B-Instruct-DPO-v0.3
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
It based on the technique described in the blog post "Refusal in LLMs is mediated by a single direction". This is an uncensored model. You can use it for any application that doesn't require alignment, like role-playing. Daredevil-8B-abliterated is the second best-performing 8B model on the Open LLM Leaderboard in terms of MMLU score (27 May 24).
译文排队中 · 当前显示来源原文
输入 文本→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
分类待补充
输入 待核验→ 输出 待核验
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
分类待补充
输入 待核验→ 输出 待核验
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
分类待补充
输入 待核验→ 输出 待核验
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
输入 文本 · 图像→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 文本 · 图像→ 输出 文本
This is an uncensored version of google/gemma-3-12b-it created with a new abliteration technique. See this article to know more about abliteration. The refusal direction is computed by comparing the residual streams between target (harmful) and baseline (harmless) samples. The hidden states of target modules (e.g., o_proj) are orthogonalized to subtract this refusal direction with a given weight factor. These weight factors follow a normal distribution with a certain spread and peak layer. Modules can be iteratively orthogonalized in batches, or the refusal direction can be accumulated to save memory. Finally, I used a hybrid evaluation with a dedicated test set to calculate the acceptance rate. This uses both a dictionary approach and NousResearch/Minos-v1. The goal is to obtain an acceptance rate >90% and still produce coherent outputs.
译文排队中 · 当前显示来源原文
输入 文本 · 图像→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 文本 · 图像→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 文本 · 图像→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
输入 文本 · 图像→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 文本 · 图像→ 输出 文本
This is an uncensored version of google/gemma-3-27b-it created with a new abliteration technique. See this article to know more about abliteration. I was playing with model weights and noticed that Gemma 3 was much more resilient to abliteration than other models like Qwen 2.5. I experimented with a few recipes to remove refusals while preserving most of the model capabilities. In the original technique, a refusal direction is computed by comparing the residual streams between target (harmful) and baseline (harmless) samples.
译文排队中 · 当前显示来源原文
输入 文本 · 图像→ 输出 文本
This is an uncensored version of google/gemma-3-27b-it created with a new abliteration technique. See this article to know more about abliteration. I was playing with model weights and noticed that Gemma 3 was much more resilient to abliteration than other models like Qwen 2.5. I experimented with a few recipes to remove refusals while preserving most of the model capabilities. In the original technique, a refusal direction is computed by comparing the residual streams between target (harmful) and baseline (harmless) samples.
译文排队中 · 当前显示来源原文
输入 文本 · 图像→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 文本 · 图像→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
输入 文本 · 图像→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
输入 文本 · 图像→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 文本 · 图像→ 输出 文本
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
分类待补充
输入 待核验→ 输出 待核验
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
输入 文本→ 输出 文本
