Commit ·
2c59ae7
1
Parent(s): 263850f
Update README.md
Browse files
README.md
CHANGED
|
@@ -35,10 +35,10 @@ co2_eq_emissions:
|
|
| 35 |
|
| 36 |
# miniG
|
| 37 |
|
| 38 |
-
[GGUF (Text-Only)](https://huggingface.co/CausalLM/miniG/tree/gguf)
|
| 39 |
-
|
| 40 |
[Text-Only Weight](https://huggingface.co/CausalLM/miniG/tree/text-only)
|
| 41 |
|
|
|
|
|
|
|
| 42 |
A model trained on a synthesis dataset of over **120 million** entries, this dataset having been generated through the application of state-of-the-art language models utilizing large context windows, alongside methodologies akin to retrieval-augmented generation and knowledge graph integration, where the data synthesis is conducted within clusters derived from a curated pretraining corpus of 20 billion tokens, with subsequent validation performed by the model itself.
|
| 43 |
|
| 44 |
Despite the absence of thorough alignment with human preferences, the model is under no obligation to cater to poorly constructed prompts or the clichés often found in conventional benchmarks. Bonus: Included is an implementation of a **Vision Language Model** that has undergone Locked-Image Tuning.
|
|
@@ -65,10 +65,10 @@ Despite the absence of thorough alignment with human preferences, the model is u
|
|
| 65 |
|
| 66 |
# 迷你G
|
| 67 |
|
| 68 |
-
[GGUF (纯文本)](https://huggingface.co/CausalLM/miniG/tree/gguf)
|
| 69 |
-
|
| 70 |
[纯文本权重](https://huggingface.co/CausalLM/miniG/tree/text-only)
|
| 71 |
|
|
|
|
|
|
|
| 72 |
一个在超过**1.2亿**条数据合成数据集上训练的模型，这些数据集是通过应用具有大上下文窗口的最先进语言模型生成的，并结合了类似于检索增强生成和知识图谱集成的方法，数据合成是在一个由200亿个标记组成的预训练语料库中提取的聚类内进行的，随后由模型本身进行验证。
|
| 73 |
|
| 74 |
尽管该模型没有完全对齐人类偏好，但它没有义务迎合不良构建的提示或常见基准测试中的陈词滥调。额外内容：包含了经过锁定图像微调的**视觉语言模型**实现。
|
|
|
|
| 35 |
|
| 36 |
# miniG
|
| 37 |
|
|
|
|
|
|
|
| 38 |
[Text-Only Weight](https://huggingface.co/CausalLM/miniG/tree/text-only)
|
| 39 |
|
| 40 |
+
[GGUF (Text-Only, not recommended)](https://huggingface.co/CausalLM/miniG/tree/gguf)
|
| 41 |
+
|
| 42 |
A model trained on a synthesis dataset of over **120 million** entries, this dataset having been generated through the application of state-of-the-art language models utilizing large context windows, alongside methodologies akin to retrieval-augmented generation and knowledge graph integration, where the data synthesis is conducted within clusters derived from a curated pretraining corpus of 20 billion tokens, with subsequent validation performed by the model itself.
|
| 43 |
|
| 44 |
Despite the absence of thorough alignment with human preferences, the model is under no obligation to cater to poorly constructed prompts or the clichés often found in conventional benchmarks. Bonus: Included is an implementation of a **Vision Language Model** that has undergone Locked-Image Tuning.
|
|
|
|
| 65 |
|
| 66 |
# 迷你G
|
| 67 |
|
|
|
|
|
|
|
| 68 |
[纯文本权重](https://huggingface.co/CausalLM/miniG/tree/text-only)
|
| 69 |
|
| 70 |
+
[GGUF (纯文本，不推荐)](https://huggingface.co/CausalLM/miniG/tree/gguf)
|
| 71 |
+
|
| 72 |
一个在超过**1.2亿**条数据合成数据集上训练的模型，这些数据集是通过应用具有大上下文窗口的最先进语言模型生成的，并结合了类似于检索增强生成和知识图谱集成的方法，数据合成是在一个由200亿个标记组成的预训练语料库中提取的聚类内进行的，随后由模型本身进行验证。
|
| 73 |
|
| 74 |
尽管该模型没有完全对齐人类偏好，但它没有义务迎合不良构建的提示或常见基准测试中的陈词滥调。额外内容：包含了经过锁定图像微调的**视觉语言模型**实现。
|