bartowski commited on
Commit
bb73768
1 Parent(s): adf452c

measurement.json

Browse files
Files changed (2) hide show
  1. README.md +68 -0
  2. measurement.json +0 -0
README.md ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model:
3
+ - Test157t/Pasta-Sea-7b-128k
4
+ - Locutusque/Hercules-2.0-Mistral-7B
5
+ library_name: transformers
6
+ tags:
7
+ - mergekit
8
+ - merge
9
+ quantized_by: bartowski
10
+ pipeline_tag: text-generation
11
+ ---
12
+
13
+ ## Exllama v2 Quantizations of HerculeanSea-7b-128k
14
+
15
+ Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.13">turboderp's ExLlamaV2 v0.0.13</a> for quantization.
16
+
17
+ <b>The "main" branch only contains the measurement.json, download one of the other branches for the model (see below)</b>
18
+
19
+ Each branch contains an individual bits per weight, with the main one containing only the meaurement.json for further conversions.
20
+
21
+ Original model: https://huggingface.co/Test157t/HerculeanSea-7b-128k
22
+
23
+ | Branch | Bits | lm_head bits | VRAM (4k) | VRAM (16k) | VRAM (32k) | Description |
24
+ | ----- | ---- | ------- | ------ | ------ | ------ | ------------ |
25
+ | [8_0](https://huggingface.co/bartowski/HerculeanSea-7b-128k-exl2/tree/8_0) | 8.0 | 8.0 | 8.4 GB | 9.8 GB | 11.8 GB | Maximum quality that ExLlamaV2 can produce, near unquantized performance. |
26
+ | [6_5](https://huggingface.co/bartowski/HerculeanSea-7b-128k-exl2/tree/6_5) | 6.5 | 8.0 | 7.2 GB | 8.6 GB | 10.6 GB | Very similar to 8.0, good tradeoff of size vs performance, **recommended**. |
27
+ | [5_0](https://huggingface.co/bartowski/HerculeanSea-7b-128k-exl2/tree/5_0) | 5.0 | 6.0 | 6.0 GB | 7.4 GB | 9.4 GB | Slightly lower quality vs 6.5, but usable on 8GB cards. |
28
+ | [4_25](https://huggingface.co/bartowski/HerculeanSea-7b-128k-exl2/tree/4_25) | 4.25 | 6.0 | 5.3 GB | 6.7 GB | 8.7 GB | GPTQ equivalent bits per weight, slightly higher quality. |
29
+ | [3_5](https://huggingface.co/bartowski/HerculeanSea-7b-128k-exl2/tree/3_5) | 3.5 | 6.0 | 4.7 GB | 6.1 GB | 8.1 GB | Lower quality, only use if you have to. |
30
+
31
+ ## Download instructions
32
+
33
+ With git:
34
+
35
+ ```shell
36
+ git clone --single-branch --branch 6_5 https://huggingface.co/bartowski/HerculeanSea-7b-128k-exl2 HerculeanSea-7b-128k-exl2-6_5
37
+ ```
38
+
39
+ With huggingface hub (credit to TheBloke for instructions):
40
+
41
+ ```shell
42
+ pip3 install huggingface-hub
43
+ ```
44
+
45
+ To download the `main` (only useful if you only care about measurement.json) branch to a folder called `HerculeanSea-7b-128k-exl2`:
46
+
47
+ ```shell
48
+ mkdir HerculeanSea-7b-128k-exl2
49
+ huggingface-cli download bartowski/HerculeanSea-7b-128k-exl2 --local-dir HerculeanSea-7b-128k-exl2 --local-dir-use-symlinks False
50
+ ```
51
+
52
+ To download from a different branch, add the `--revision` parameter:
53
+
54
+ Linux:
55
+
56
+ ```shell
57
+ mkdir HerculeanSea-7b-128k-exl2-6_5
58
+ huggingface-cli download bartowski/HerculeanSea-7b-128k-exl2 --revision 6_5 --local-dir HerculeanSea-7b-128k-exl2-6_5 --local-dir-use-symlinks False
59
+ ```
60
+
61
+ Windows (which apparently doesn't like _ in folders sometimes?):
62
+
63
+ ```shell
64
+ mkdir HerculeanSea-7b-128k-exl2-6.5
65
+ huggingface-cli download bartowski/HerculeanSea-7b-128k-exl2 --revision 6_5 --local-dir HerculeanSea-7b-128k-exl2-6.5 --local-dir-use-symlinks False
66
+ ```
67
+
68
+ Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski
measurement.json ADDED
The diff for this file is too large to render. See raw diff