bartowski commited on
Commit
2bb78fb
1 Parent(s): 9b18b46

measurement.json

Browse files
Files changed (2) hide show
  1. README.md +76 -0
  2. measurement.json +0 -0
README.md ADDED
@@ -0,0 +1,76 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ datasets:
4
+ - Azure99/blossom-chat-v3
5
+ - Azure99/blossom-math-v4
6
+ - Azure99/blossom-wizard-v3
7
+ - Azure99/blossom-orca-v3
8
+ language:
9
+ - zh
10
+ - en
11
+ quantized_by: bartowski
12
+ pipeline_tag: text-generation
13
+ ---
14
+
15
+ ## Exllama v2 Quantizations of blossom-v5-9b
16
+
17
+ Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.15">turboderp's ExLlamaV2 v0.0.15</a> for quantization.
18
+
19
+ ## The "main" branch only contains the measurement.json, download one of the other branches for the model (see below)
20
+
21
+ Each branch contains an individual bits per weight, with the main one containing only the meaurement.json for further conversions.
22
+
23
+ Conversion was done using the default calibration dataset.
24
+
25
+ Default arguments used except when the bits per weight is above 6.0, at that point the lm_head layer is quantized at 8 bits per weight instead of the default 6.
26
+
27
+ Original model: https://huggingface.co/Azure99/blossom-v5-9b
28
+
29
+
30
+ <a href="https://huggingface.co/bartowski/blossom-v5-9b-exl2/tree/8_0">8.0 bits per weight</a>
31
+
32
+ <a href="https://huggingface.co/bartowski/blossom-v5-9b-exl2/tree/6_5">6.5 bits per weight</a>
33
+
34
+ <a href="https://huggingface.co/bartowski/blossom-v5-9b-exl2/tree/5_0">5.0 bits per weight</a>
35
+
36
+ <a href="https://huggingface.co/bartowski/blossom-v5-9b-exl2/tree/4_25">4.25 bits per weight</a>
37
+
38
+ <a href="https://huggingface.co/bartowski/blossom-v5-9b-exl2/tree/3_5">3.5 bits per weight</a>
39
+
40
+
41
+ ## Download instructions
42
+
43
+ With git:
44
+
45
+ ```shell
46
+ git clone --single-branch --branch 6_5 https://huggingface.co/bartowski/blossom-v5-9b-exl2
47
+ ```
48
+
49
+ With huggingface hub (credit to TheBloke for instructions):
50
+
51
+ ```shell
52
+ pip3 install huggingface-hub
53
+ ```
54
+
55
+ To download the `main` (only useful if you only care about measurement.json) branch to a folder called `blossom-v5-9b-exl2`:
56
+
57
+ ```shell
58
+ mkdir blossom-v5-9b-exl2
59
+ huggingface-cli download bartowski/blossom-v5-9b-exl2 --local-dir blossom-v5-9b-exl2 --local-dir-use-symlinks False
60
+ ```
61
+
62
+ To download from a different branch, add the `--revision` parameter:
63
+
64
+ Linux:
65
+
66
+ ```shell
67
+ mkdir blossom-v5-9b-exl2-6_5
68
+ huggingface-cli download bartowski/blossom-v5-9b-exl2 --revision 6_5 --local-dir blossom-v5-9b-exl2-6_5 --local-dir-use-symlinks False
69
+ ```
70
+
71
+ Windows (which apparently doesn't like _ in folders sometimes?):
72
+
73
+ ```shell
74
+ mkdir blossom-v5-9b-exl2-6.5
75
+ huggingface-cli download bartowski/blossom-v5-9b-exl2 --revision 6_5 --local-dir blossom-v5-9b-exl2-6.5 --local-dir-use-symlinks False
76
+ ```
measurement.json ADDED
The diff for this file is too large to render. See raw diff