mradermacher commited on
Commit
42d9c38
·
verified ·
1 Parent(s): 8339620

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +106 -0
README.md CHANGED
@@ -1,6 +1,112 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  <!-- ### quantize_version: 2 -->
2
  <!-- ### output_tensor_quantised: 1 -->
3
  <!-- ### convert_type: hf -->
4
  <!-- ### vocab_type: -->
5
  <!-- ### tags: -->
6
  static quants of https://huggingface.co/Weyaxi/Einstein-v4-7B
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Weyaxi/Einstein-v4-7B
3
+ datasets:
4
+ - allenai/ai2_arc
5
+ - camel-ai/physics
6
+ - camel-ai/chemistry
7
+ - camel-ai/biology
8
+ - camel-ai/math
9
+ - metaeval/reclor
10
+ - openbookqa
11
+ - mandyyyyii/scibench
12
+ - derek-thomas/ScienceQA
13
+ - TIGER-Lab/ScienceEval
14
+ - jondurbin/airoboros-3.2
15
+ - LDJnr/Capybara
16
+ - Cot-Alpaca-GPT4-From-OpenHermes-2.5
17
+ - STEM-AI-mtl/Electrical-engineering
18
+ - knowrohit07/saraswati-stem
19
+ - sablo/oasst2_curated
20
+ - glaiveai/glaive-code-assistant
21
+ - lmsys/lmsys-chat-1m
22
+ - TIGER-Lab/MathInstruct
23
+ - bigbio/med_qa
24
+ - meta-math/MetaMathQA-40K
25
+ - openbookqa
26
+ - piqa
27
+ - metaeval/reclor
28
+ - derek-thomas/ScienceQA
29
+ - scibench
30
+ - sciq
31
+ - Open-Orca/SlimOrca
32
+ - migtissera/Synthia-v1.3
33
+ - TIGER-Lab/ScienceEval
34
+ language:
35
+ - en
36
+ library_name: transformers
37
+ license: other
38
+ quantized_by: mradermacher
39
+ tags:
40
+ - axolotl
41
+ - generated_from_trainer
42
+ - Mistral
43
+ - instruct
44
+ - finetune
45
+ - chatml
46
+ - gpt4
47
+ - synthetic data
48
+ - science
49
+ - physics
50
+ - chemistry
51
+ - biology
52
+ - math
53
+ ---
54
+ ## About
55
+
56
  <!-- ### quantize_version: 2 -->
57
  <!-- ### output_tensor_quantised: 1 -->
58
  <!-- ### convert_type: hf -->
59
  <!-- ### vocab_type: -->
60
  <!-- ### tags: -->
61
  static quants of https://huggingface.co/Weyaxi/Einstein-v4-7B
62
+
63
+ <!-- provided-files -->
64
+ weighted/imatrix quants are available at https://huggingface.co/mradermacher/Einstein-v4-7B-i1-GGUF
65
+ ## Usage
66
+
67
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
68
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
69
+ more details, including on how to concatenate multi-part files.
70
+
71
+ ## Provided Quants
72
+
73
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
74
+
75
+ | Link | Type | Size/GB | Notes |
76
+ |:-----|:-----|--------:|:------|
77
+ | [GGUF](https://huggingface.co/mradermacher/Einstein-v4-7B-GGUF/resolve/main/Einstein-v4-7B.Q2_K.gguf) | Q2_K | 2.8 | |
78
+ | [GGUF](https://huggingface.co/mradermacher/Einstein-v4-7B-GGUF/resolve/main/Einstein-v4-7B.IQ3_XS.gguf) | IQ3_XS | 3.1 | |
79
+ | [GGUF](https://huggingface.co/mradermacher/Einstein-v4-7B-GGUF/resolve/main/Einstein-v4-7B.Q3_K_S.gguf) | Q3_K_S | 3.3 | |
80
+ | [GGUF](https://huggingface.co/mradermacher/Einstein-v4-7B-GGUF/resolve/main/Einstein-v4-7B.IQ3_S.gguf) | IQ3_S | 3.3 | beats Q3_K* |
81
+ | [GGUF](https://huggingface.co/mradermacher/Einstein-v4-7B-GGUF/resolve/main/Einstein-v4-7B.IQ3_M.gguf) | IQ3_M | 3.4 | |
82
+ | [GGUF](https://huggingface.co/mradermacher/Einstein-v4-7B-GGUF/resolve/main/Einstein-v4-7B.Q3_K_M.gguf) | Q3_K_M | 3.6 | lower quality |
83
+ | [GGUF](https://huggingface.co/mradermacher/Einstein-v4-7B-GGUF/resolve/main/Einstein-v4-7B.Q3_K_L.gguf) | Q3_K_L | 3.9 | |
84
+ | [GGUF](https://huggingface.co/mradermacher/Einstein-v4-7B-GGUF/resolve/main/Einstein-v4-7B.IQ4_XS.gguf) | IQ4_XS | 4.0 | |
85
+ | [GGUF](https://huggingface.co/mradermacher/Einstein-v4-7B-GGUF/resolve/main/Einstein-v4-7B.Q4_K_S.gguf) | Q4_K_S | 4.2 | fast, recommended |
86
+ | [GGUF](https://huggingface.co/mradermacher/Einstein-v4-7B-GGUF/resolve/main/Einstein-v4-7B.Q4_K_M.gguf) | Q4_K_M | 4.5 | fast, recommended |
87
+ | [GGUF](https://huggingface.co/mradermacher/Einstein-v4-7B-GGUF/resolve/main/Einstein-v4-7B.Q5_K_S.gguf) | Q5_K_S | 5.1 | |
88
+ | [GGUF](https://huggingface.co/mradermacher/Einstein-v4-7B-GGUF/resolve/main/Einstein-v4-7B.Q5_K_M.gguf) | Q5_K_M | 5.2 | |
89
+ | [GGUF](https://huggingface.co/mradermacher/Einstein-v4-7B-GGUF/resolve/main/Einstein-v4-7B.Q6_K.gguf) | Q6_K | 6.0 | very good quality |
90
+ | [GGUF](https://huggingface.co/mradermacher/Einstein-v4-7B-GGUF/resolve/main/Einstein-v4-7B.Q8_0.gguf) | Q8_0 | 7.8 | fast, best quality |
91
+ | [GGUF](https://huggingface.co/mradermacher/Einstein-v4-7B-GGUF/resolve/main/Einstein-v4-7B.f16.gguf) | f16 | 14.6 | 16 bpw, overkill |
92
+
93
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
94
+ types (lower is better):
95
+
96
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
97
+
98
+ And here are Artefact2's thoughts on the matter:
99
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
100
+
101
+ ## FAQ / Model Request
102
+
103
+ See https://huggingface.co/mradermacher/model_requests for some answers to
104
+ questions you might have and/or if you want some other model quantized.
105
+
106
+ ## Thanks
107
+
108
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
109
+ me use its servers and providing upgrades to my workstation to enable
110
+ this work in my free time.
111
+
112
+ <!-- end -->