Upload folder using huggingface_hub
Browse files- README.md +1 -2
- config.json +3 -4
- configuration.json +1 -0
- llm.mnn +3 -0
- llm.mnn.json +3 -0
- llm.mnn.weight +3 -0
- llm_config.json +2 -3
- qwen-7b-int4.mnn +0 -0
- tokenizer.txt +2 -2
README.md
CHANGED
@@ -9,5 +9,4 @@ tags:
|
|
9 |
# Qwen-7B-Chat-MNN
|
10 |
|
11 |
## Introduction
|
12 |
-
|
13 |
-
This model is a 4-bit quantized version of the MNN model exported from Qwen-7B-Chat using [llm-export](https://github.com/wangzhaode/llm-export).
|
|
|
9 |
# Qwen-7B-Chat-MNN
|
10 |
|
11 |
## Introduction
|
12 |
+
This model is a 4-bit quantized version of the MNN model exported from [Qwen-7B-Chat](https://huggingface.co/Qwen/Qwen-7B-Chat) using [llm-export](https://github.com/wangzhaode/llm-export).
|
|
config.json
CHANGED
@@ -1,9 +1,8 @@
|
|
1 |
{
|
2 |
-
"llm_model": "
|
3 |
-
"llm_weight": "
|
4 |
-
|
5 |
"backend_type": "cpu",
|
6 |
"thread_num": 4,
|
7 |
"precision": "low",
|
8 |
"memory": "low"
|
9 |
-
}
|
|
|
1 |
{
|
2 |
+
"llm_model": "llm.mnn",
|
3 |
+
"llm_weight": "llm.mnn.weight",
|
|
|
4 |
"backend_type": "cpu",
|
5 |
"thread_num": 4,
|
6 |
"precision": "low",
|
7 |
"memory": "low"
|
8 |
+
}
|
configuration.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"framework":"other","task":"text-generation"}
|
llm.mnn
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:17928480eac907e08dff44151067a931f3697430337db93a92cb47d2e1c481bd
|
3 |
+
size 2630448
|
llm.mnn.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:35017daa90de6a651f2e4bca1635f247a55d2f4706cff7f51cbcbe077dbbca38
|
3 |
+
size 16696560
|
llm.mnn.weight
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:52504f59db03e0508c9d9cb5d5aca476378a511007807c8f667b72d749ca5bf3
|
3 |
+
size 3994391386
|
llm_config.json
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
{
|
2 |
"hidden_size": 4096,
|
3 |
"layer_nums": 32,
|
4 |
-
"attention_mask": "
|
5 |
"key_value_shape": [
|
6 |
2,
|
7 |
1,
|
@@ -10,6 +10,5 @@
|
|
10 |
128
|
11 |
],
|
12 |
"prompt_template": "\n<|im_start|>user\n%s<|im_end|>\n<|im_start|>assistant\n",
|
13 |
-
"is_visual": false
|
14 |
-
"is_single": true
|
15 |
}
|
|
|
1 |
{
|
2 |
"hidden_size": 4096,
|
3 |
"layer_nums": 32,
|
4 |
+
"attention_mask": "float",
|
5 |
"key_value_shape": [
|
6 |
2,
|
7 |
1,
|
|
|
10 |
128
|
11 |
],
|
12 |
"prompt_template": "\n<|im_start|>user\n%s<|im_end|>\n<|im_start|>assistant\n",
|
13 |
+
"is_visual": false
|
|
|
14 |
}
|
qwen-7b-int4.mnn
CHANGED
Binary files a/qwen-7b-int4.mnn and b/qwen-7b-int4.mnn differ
|
|
tokenizer.txt
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
430 1
|
2 |
-
208
|
3 |
-
151643 151644 151645 151646 151647 151648 151649 151650 151651 151652 151653 151654 151655 151656 151657 151658 151659 151660 151661 151662 151663 151664 151665 151666 151667 151668 151669 151670 151671 151672 151673 151674 151675 151676 151677 151678 151679 151680 151681 151682 151683 151684 151685 151686 151687 151688 151689 151690 151691 151692 151693 151694 151695 151696 151697 151698 151699 151700 151701 151702 151703 151704 151705 151706 151707 151708 151709 151710 151711 151712 151713 151714 151715 151716 151717 151718 151719 151720 151721 151722 151723 151724 151725 151726 151727 151728 151729 151730 151731 151732 151733 151734 151735 151736 151737 151738 151739 151740 151741 151742 151743 151744 151745 151746 151747 151748 151749 151750 151751 151752 151753 151754 151755 151756 151757 151758 151759 151760 151761 151762 151763 151764 151765 151766 151767 151768 151769 151770 151771 151772 151773 151774 151775 151776 151777 151778 151779 151780 151781 151782 151783 151784 151785 151786 151787 151788 151789 151790 151791 151792 151793 151794 151795 151796 151797 151798 151799 151800 151801 151802 151803 151804 151805 151806 151807 151808 151809 151810 151811 151812 151813 151814 151815 151816 151817 151818 151819 151820 151821 151822 151823 151824 151825 151826 151827 151828 151829 151830 151831 151832 151833 151834 151835 151836 151837 151838 151839 151840 151841 151842 151843 151844 151845 151846 151847 151848 151849 151850 151645
|
4 |
151851
|
5 |
IQ==
|
6 |
Ig==
|
|
|
1 |
430 1
|
2 |
+
208 2 0
|
3 |
+
151643 151644 151645 151646 151647 151648 151649 151650 151651 151652 151653 151654 151655 151656 151657 151658 151659 151660 151661 151662 151663 151664 151665 151666 151667 151668 151669 151670 151671 151672 151673 151674 151675 151676 151677 151678 151679 151680 151681 151682 151683 151684 151685 151686 151687 151688 151689 151690 151691 151692 151693 151694 151695 151696 151697 151698 151699 151700 151701 151702 151703 151704 151705 151706 151707 151708 151709 151710 151711 151712 151713 151714 151715 151716 151717 151718 151719 151720 151721 151722 151723 151724 151725 151726 151727 151728 151729 151730 151731 151732 151733 151734 151735 151736 151737 151738 151739 151740 151741 151742 151743 151744 151745 151746 151747 151748 151749 151750 151751 151752 151753 151754 151755 151756 151757 151758 151759 151760 151761 151762 151763 151764 151765 151766 151767 151768 151769 151770 151771 151772 151773 151774 151775 151776 151777 151778 151779 151780 151781 151782 151783 151784 151785 151786 151787 151788 151789 151790 151791 151792 151793 151794 151795 151796 151797 151798 151799 151800 151801 151802 151803 151804 151805 151806 151807 151808 151809 151810 151811 151812 151813 151814 151815 151816 151817 151818 151819 151820 151821 151822 151823 151824 151825 151826 151827 151828 151829 151830 151831 151832 151833 151834 151835 151836 151837 151838 151839 151840 151841 151842 151843 151844 151845 151846 151847 151848 151849 151850 151643 151645
|
4 |
151851
|
5 |
IQ==
|
6 |
Ig==
|