bartowski commited on
Commit
3969537
1 Parent(s): e5b1209

measurement.json

Browse files
Files changed (2) hide show
  1. README.md +57 -0
  2. measurement.json +0 -0
README.md ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ pipeline_tag: text-generation
3
+ license: other
4
+ quantized_by: bartowski
5
+ ---
6
+
7
+ ## Exllama v2 Quantizations of internlm2-20b-llama
8
+
9
+ Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.11">turboderp's ExLlamaV2 v0.0.11</a> for quantization.
10
+
11
+ # The "main" branch only contains the measurement.json, download one of the other branches for the model (see below)
12
+
13
+ Each branch contains an individual bits per weight, with the main one containing only the meaurement.json for further conversions.
14
+
15
+ Conversion was done using the default calibration dataset.
16
+
17
+ 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.
18
+
19
+ Original model: https://huggingface.co/internlm/internlm2-20b
20
+
21
+
22
+
23
+ <a href="https://huggingface.co/bartowski/internlm2-20b-llama-exl2/tree/6_5">6.5 bits per weight</a>
24
+
25
+ <a href="https://huggingface.co/bartowski/internlm2-20b-llama-exl2/tree/4_5">4.5 bits per weight</a>
26
+
27
+ <a href="https://huggingface.co/bartowski/internlm2-20b-llama-exl2/tree/3_5">3.5 bits per weight</a>
28
+
29
+ <a href="https://huggingface.co/bartowski/internlm2-20b-llama-exl2/tree/3_0">3.0 bits per weight</a>
30
+
31
+ ## Download instructions
32
+
33
+ With git:
34
+
35
+ ```shell
36
+ git clone --single-branch --branch 4_0 https://huggingface.co/bartowski/internlm2-20b-llama-exl2
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 `internlm2-20b-llama-exl2`:
46
+
47
+ ```shell
48
+ mkdir internlm2-20b-llama-exl2
49
+ huggingface-cli download bartowski/internlm2-20b-llama-exl2 --local-dir internlm2-20b-llama-exl2 --local-dir-use-symlinks False
50
+ ```
51
+
52
+ To download from a different branch, add the `--revision` parameter:
53
+
54
+ ```shell
55
+ mkdir internlm2-20b-llama-exl2
56
+ huggingface-cli download bartowski/internlm2-20b-llama-exl2 --revision 4_0 --local-dir internlm2-20b-llama-exl2 --local-dir-use-symlinks False
57
+ ```
measurement.json ADDED
The diff for this file is too large to render. See raw diff