mayank-mishra commited on
Commit
24bfe40
1 Parent(s): e8b446c
Files changed (2) hide show
  1. README.md +21 -0
  2. model.pt +3 -0
README.md ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: openrail
3
+ ---
4
+
5
+ # GPTQ-for-StarCoder
6
+ Visit [GPTQ-for-SantaCoder](https://github.com/mayank31398/GPTQ-for-SantaCoder) for instructions on how to use the model weights here.
7
+ If you want 4-bit weights, visit [starcoderbase-GPTQ-4bit-128g](https://huggingface.co/mayank31398/starcoderbase-GPTQ-4bit-128g).
8
+
9
+ ## Results
10
+ | StarCoderBase | Bits | group-size | memory(MiB) | wikitext2 | ptb | c4 | stack | checkpoint size(MB) |
11
+ | -------------------------------------------------- | ---- | ---------- | ----------- | --------- | ---------- | ---------- | ---------- | ------------------- |
12
+ | FP32 | 32 | - | | 10.172 | 15.756 | 12.736 | 1.692 | 59195 |
13
+ | BF16 | 16 | - | | 10.173 | 15.765 | 12.745 | 1.692 | 29597 |
14
+ | [GPTQ](https://arxiv.org/abs/2210.17323) | 8 | 128 | | 10.174 | 15.767 | 12.739 | 1.692 | 16163 |
15
+ | [GPTQ](https://arxiv.org/abs/2210.17323) | 4 | 128 | | 10.418 | 16.190 | 13.018 | 1.708 | 8877 |
16
+
17
+ # License
18
+ The model is licenses under the CodeML Open RAIL-M v0.1 license. You can find the full license [here](https://huggingface.co/spaces/bigcode/license).
19
+
20
+ # Acknowledgements
21
+ Thanks to everyone in BigCode who worked so hard to create these code models.
model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43389afaa1a894a2c3df5c4a54bdac5a46870b3ea38a7acec88a77f8b435f27f
3
+ size 16948589225