Hugging Face's logo

MECHUK
/
embeddinggemma-rus-32768

Sentence Similarity
GGUF
Russian
llama.cpp
trimmed
quantized
Model card Files
xet
 Community
MECHUK commited on
Commit
5abf646
·
verified ·
1 Parent(s): 012bec3

feat: add GGUF Q8_0 + F32 quantizations of embeddinggemma-rus-32768

Browse files

Converted with llama.cpp c1a1c8ee; includes SHA256SUMS, conversion-report, smoke-embedding. README mirrors upstream alphaedge-ai/embeddinggemma-rus-32768, adapted for GGUF/llama.cpp usage.

Files changed (7) hide show
  1. .gitattributes +1 -35
  2. README.md +108 -0
  3. SHA256SUMS +3 -0
  4. conversion-report.txt +66 -0
  5. embeddinggemma-rus-32768-F32.gguf +3 -0
  6. embeddinggemma-rus-32768-Q8_0.gguf +3 -0
  7. smoke-embedding.json +11 -0
.gitattributes CHANGED
@@ -1,35 +1 @@
1
- *.7z filter=lfs diff=lfs merge=lfs -text
2
- *.arrow filter=lfs diff=lfs merge=lfs -text
3
- *.bin filter=lfs diff=lfs merge=lfs -text
4
- *.bz2 filter=lfs diff=lfs merge=lfs -text
5
- *.ckpt filter=lfs diff=lfs merge=lfs -text
6
- *.ftz filter=lfs diff=lfs merge=lfs -text
7
- *.gz filter=lfs diff=lfs merge=lfs -text
8
- *.h5 filter=lfs diff=lfs merge=lfs -text
9
- *.joblib filter=lfs diff=lfs merge=lfs -text
10
- *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
- *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
- *.model filter=lfs diff=lfs merge=lfs -text
13
- *.msgpack filter=lfs diff=lfs merge=lfs -text
14
- *.npy filter=lfs diff=lfs merge=lfs -text
15
- *.npz filter=lfs diff=lfs merge=lfs -text
16
- *.onnx filter=lfs diff=lfs merge=lfs -text
17
- *.ot filter=lfs diff=lfs merge=lfs -text
18
- *.parquet filter=lfs diff=lfs merge=lfs -text
19
- *.pb filter=lfs diff=lfs merge=lfs -text
20
- *.pickle filter=lfs diff=lfs merge=lfs -text
21
- *.pkl filter=lfs diff=lfs merge=lfs -text
22
- *.pt filter=lfs diff=lfs merge=lfs -text
23
- *.pth filter=lfs diff=lfs merge=lfs -text
24
- *.rar filter=lfs diff=lfs merge=lfs -text
25
- *.safetensors filter=lfs diff=lfs merge=lfs -text
26
- saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
- *.tar.* filter=lfs diff=lfs merge=lfs -text
28
- *.tar filter=lfs diff=lfs merge=lfs -text
29
- *.tflite filter=lfs diff=lfs merge=lfs -text
30
- *.tgz filter=lfs diff=lfs merge=lfs -text
31
- *.wasm filter=lfs diff=lfs merge=lfs -text
32
- *.xz filter=lfs diff=lfs merge=lfs -text
33
- *.zip filter=lfs diff=lfs merge=lfs -text
34
- *.zst filter=lfs diff=lfs merge=lfs -text
35
- *tfevents* filter=lfs diff=lfs merge=lfs -text
 
1
+ *.gguf filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
README.md ADDED
@@ -0,0 +1,108 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ pipeline_tag: sentence-similarity
3
+ language: rus
4
+ license: gemma
5
+ tags:
6
+ - gguf
7
+ - llama.cpp
8
+ - trimmed
9
+ - quantized
10
+ library_name: gguf
11
+ base_model: alphaedge-ai/embeddinggemma-rus-32768
12
+ base_model_relation: quantized
13
+ datasets:
14
+ - lbourdois/fineweb-2-trimming
15
+ ---
16
+
17
+ # embeddinggemma-rus-32768 (GGUF)
18
+
19
+ GGUF quantizations of [alphaedge-ai/embeddinggemma-rus-32768](https://huggingface.co/alphaedge-ai/embeddinggemma-rus-32768), which is a **57.27% smaller** version of [google/embeddinggemma-300m](https://huggingface.co/google/embeddinggemma-300m) optimized for Russian language via vocabulary size reduction using the [trimming](https://huggingface.co/blog/lbourdois/introduction-to-trimming) method.
20
+
21
+ ## Model Statistics
22
+ | Metric | Original | Trimmed | Reduction |
23
+ |--------|----------|---------|-----------|
24
+ | **Vocabulary size** | 262,144 tokens | 32,768 tokens | **87.50%** |
25
+ | **Model size** | 307,581,696 params | 131,420,928 params | **57.27%** |
26
+
27
+ ![image](https://raw.githubusercontent.com/lbourdois/blog/refs/heads/master/assets/images/Trimming/embeddingemma-32768.png)
28
+
29
+ ## GGUF Quantizations
30
+
31
+ | File | Type | Size |
32
+ |------|------|------|
33
+ | `embeddinggemma-rus-32768-Q8_0.gguf` | Q8_0 (8-bit) | 136 MB |
34
+ | `embeddinggemma-rus-32768-F32.gguf` | F32 (lossless reference) | 503 MB |
35
+
36
+ Integrity checksums are in [`SHA256SUMS`](./SHA256SUMS). **Q8_0 is the recommended default**; F32 is provided as a lossless reference equivalent to the source safetensors.
37
+
38
+ ## Conversion
39
+
40
+ Converted with [llama.cpp](https://github.com/ggml-org/llama.cpp) (commit `c1a1c8ee`). Full provenance — build environment, downloaded source files, exact output sizes and checksums — is recorded in [`conversion-report.txt`](./conversion-report.txt).
41
+
42
+ Two non-obvious steps were required for a correct conversion of this trimmed model:
43
+
44
+ 1. **Tokenizer registry patch.** This trimmed model ships only `tokenizer.json` (Gemma SPM-style BPE) and no `tokenizer.model` (SentencePiece). The Gemma3 HF→GGUF converter only takes the SentencePiece path when `tokenizer.model` exists, so the model's tokenizer `chkhsh` (`b847c511…`) was registered as the `gemma4` pre-type (`get_vocab_base_pre()`) in `conversion/base.py`. This maps the SPM-style BPE (normalizer ` `→`▁`, BPE over whole text, ByteFallback on raw UTF-8) correctly.
45
+ 2. **`--sentence-transformers-dense-modules`.** EmbeddingGemma has `2_Dense`/`3_Dense` projection layers; without this flag they are silently dropped and embeddings drift from the SentenceTransformers baseline.
46
+
47
+ A functional smoke test (`llama-embedding`, L2-normalized, OpenAI-style JSON output) is included as [`smoke-embedding.json`](./smoke-embedding.json).
48
+
49
+ ## Usage
50
+
51
+ With `llama-server` (OpenAI-compatible embeddings endpoint):
52
+
53
+ ```bash
54
+ llama-server \
55
+ -m embeddinggemma-rus-32768-Q8_0.gguf \
56
+ --embeddings --host 0.0.0.0 --port 8080
57
+ ```
58
+
59
+ ```bash
60
+ curl http://localhost:8080/v1/embeddings \
61
+ -H "Content-Type: application/json" \
62
+ -d '{"input": "task: search result | query: тестовый русский запрос", "model": "embeddinggemma-rus-32768"}'
63
+ ```
64
+
65
+ Or directly with `llama-embedding`:
66
+
67
+ ```bash
68
+ llama-embedding \
69
+ -m embeddinggemma-rus-32768-Q8_0.gguf \
70
+ --embd-output-format json --embd-normalize 2 \
71
+ -p "task: search result | query: тестовый русский запрос"
72
+ ```
73
+
74
+ > EmbeddingGemma is instruction-tuned; prefix inputs with `task: <task> | query: <text>` (e.g. `task: search result`, `task: title`, `task: web query`).
75
+
76
+ ## Mining Dataset Statistics
77
+ - **Number of texts used for mining**: 200,000 texts
78
+ - **Dataset**: [lbourdois/fineweb-2-trimming](https://huggingface.co/datasets/lbourdois/fineweb-2-trimming)
79
+
80
+ ## Citations
81
+
82
+ #### EmbeddingGemma
83
+ ```
84
+ @misc{vera2025embeddinggemmapowerfullightweighttext,
85
+ title={EmbeddingGemma: Powerful and Lightweight Text Representations},
86
+ author={Henrique Schechter Vera and Sahil Dua and Biao Zhang and Daniel Salz and Ryan Mullins and Sindhu Raghuram Panyam and Sara Smoot and Iftekhar Naim and Joe Zou and Feiyang Chen and Daniel Cer and Alice Lisak and Min Choi and Lucas Gonzalez and Omar Sanseviero and Glenn Cameron and Ian Ballantyne and Kat Black and Kaifeng Chen and Weiyi Wang and Zhe Li and Gus Martins and Jinhyuk Lee and Mark Sherwood and Juyeong Ji and Renjie Wu and Jingxiao Zheng and Jyotinder Singh and Abheesht Sharma and Divyashree Sreepathihalli and Aashi Jain and Adham Elarabawy and AJ Co and Andreas Doumanoglou and Babak Samari and Ben Hora and Brian Potetz and Dahun Kim and Enrique Alfonseca and Fedor Moiseev and Feng Han and Frank Palma Gomez and Gustavo Hernández Ábrego and Hesen Zhang and Hui Hui and Jay Han and Karan Gill and Ke Chen and Koert Chen and Madhuri Shanbhogue and Michael Boratko and Paul Suganthan and Sai Meher Karthik Duddu and Sandeep Mariserla and Setareh Ariafar and Shanfeng Zhang and Shijie Zhang and Simon Baumgartner and Sonam Goenka and Steve Qiu and Tanmaya Dabral and Trevor Walker and Vikram Rao and Waleed Khawaja and Wenlei Zhou and Xiaoqi Ren and Ye Xia and Yichang Chen and Yi-Ting Chen and Zhe Dong and Zhongli Ding and Francesco Visin and Gaël Liu and Jiageng Zhang and Kathleen Kenealy and Michelle Casbon and Ravin Kumar and Thomas Mesnard and Zach Gleicher and Cormac Brick and Olivier Lacombe and Adam Roberts and Qin Yin and Yunhsuan Sung and Raphael Hoffmann and Tris Warkentin and Armand Joulin and Tom Duerig and Mojtaba Seyedhosseini},
87
+ year={2025},
88
+ eprint={2509.20354},
89
+ archivePrefix={arXiv},
90
+ primaryClass={cs.CL},
91
+ url={https://arxiv.org/abs/2509.20354},
92
+ }
93
+ ```
94
+
95
+ #### Trimming blog post
96
+ ```
97
+ @misc{hf_blogpost_trimming,
98
+ title={Introduction to Trimming},
99
+ author={Loïck BOURDOIS and Tom AARSEN and Bram VANROY and Christopher AKIKI and Woojun JUNG and Manuel ROMERO and Prithiv SAKTHI},
100
+ year={2026},
101
+ url={https://huggingface.co/blog/lbourdois/introduction-to-trimming},
102
+ }
103
+ ```
104
+
105
+ ### License
106
+ This model is derived from [google/embeddinggemma-300m](https://huggingface.co/google/embeddinggemma-300m).
107
+ Use of this model is governed by the [Gemma Terms of Use](https://ai.google.dev/gemma/terms).
108
+ By using this model, you agree to the Gemma Terms of Use. This model is not affiliated with or endorsed by Google.
SHA256SUMS ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ cdc20839de6745e478b5869f5eb5f3275aee79799435446b863f8cd104c79055 embeddinggemma-rus-32768-F32.gguf
2
+ dc7c84e1d37d53344d5d5b617c4db2ebeb928ab147880708e77113062b0f7f14 embeddinggemma-rus-32768-Q8_0.gguf
3
+ a0df320b451b8d513ba8dd74d027c5de7e44dc749db33966fffec0cb87e163b4 smoke-embedding.json
conversion-report.txt ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ conversion_started_utc=2026-06-28T12:49:31Z
2
+ model_repo=alphaedge-ai/embeddinggemma-rus-32768
3
+ model_name=embeddinggemma-rus-32768
4
+ threads=12
5
+ out_dir=/out
6
+
7
+ ## Environment
8
+ Linux c86dd93cc5a3 7.0.12-201.fc44.x86_64 #1 SMP PREEMPT_DYNAMIC Thu Jun 11 01:30:16 UTC 2026 x86_64 GNU/Linux
9
+ PRETTY_NAME="Debian GNU/Linux 13 (trixie)"
10
+ NAME="Debian GNU/Linux"
11
+ VERSION_ID="13"
12
+ VERSION="13 (trixie)"
13
+ VERSION_CODENAME=trixie
14
+ DEBIAN_VERSION_FULL=13.5
15
+ ID=debian
16
+ HOME_URL="https://www.debian.org/"
17
+ SUPPORT_URL="https://www.debian.org/support"
18
+ BUG_REPORT_URL="https://bugs.debian.org/"
19
+ 16
20
+ total used free shared buff/cache available
21
+ Mem: 29Gi 13Gi 3.7Gi 109Mi 12Gi 15Gi
22
+ Swap: 14Gi 10Gi 4.1Gi
23
+ Filesystem Size Used Avail Use% Mounted on
24
+ /dev/mapper/luks-ab2e11c1-ee96-4b8f-bed1-f37c94df1f03 780G 553G 217G 72% /work
25
+ /dev/mapper/luks-ab2e11c1-ee96-4b8f-bed1-f37c94df1f03 780G 553G 217G 72% /out
26
+ Python 3.12.13
27
+ cmake version 3.31.6
28
+
29
+ CMake suite maintained and supported by Kitware (kitware.com/cmake).
30
+ 1.12.1
31
+ llama_cpp_commit=c1a1c8ee94e37ff7ba2c872e783aaf7f77e0f320
32
+ tokenizer_registry_patched=true
33
+
34
+ ## Downloaded files
35
+ /work/models/embeddinggemma-rus-32768/.cache/huggingface/.gitignore 1 bytes
36
+ /work/models/embeddinggemma-rus-32768/.gitattributes 1519 bytes
37
+ /work/models/embeddinggemma-rus-32768/1_Pooling/config.json 312 bytes
38
+ /work/models/embeddinggemma-rus-32768/2_Dense/config.json 134 bytes
39
+ /work/models/embeddinggemma-rus-32768/2_Dense/model.safetensors 9437272 bytes
40
+ /work/models/embeddinggemma-rus-32768/3_Dense/config.json 134 bytes
41
+ /work/models/embeddinggemma-rus-32768/3_Dense/model.safetensors 9437272 bytes
42
+ /work/models/embeddinggemma-rus-32768/README.md 4326 bytes
43
+ /work/models/embeddinggemma-rus-32768/config.json 1571 bytes
44
+ /work/models/embeddinggemma-rus-32768/config_sentence_transformers.json 997 bytes
45
+ /work/models/embeddinggemma-rus-32768/generation_config.json 133 bytes
46
+ /work/models/embeddinggemma-rus-32768/model.safetensors 506842664 bytes
47
+ /work/models/embeddinggemma-rus-32768/modules.json 573 bytes
48
+ /work/models/embeddinggemma-rus-32768/sentence_bert_config.json 58 bytes
49
+ /work/models/embeddinggemma-rus-32768/special_tokens_map.json 585 bytes
50
+ /work/models/embeddinggemma-rus-32768/tokenizer.json 3720836 bytes
51
+ /work/models/embeddinggemma-rus-32768/tokenizer_config.json 2273 bytes
52
+
53
+ ## Output files
54
+ total 639M
55
+ -rw-r--r--. 1 root root 288 Jun 28 12:49 SHA256SUMS
56
+ -rw-r--r--. 1 root root 2.5K Jun 28 12:49 conversion-report.txt
57
+ -rw-r--r--. 1 root root 503M Jun 28 12:49 embeddinggemma-rus-32768-F32.gguf
58
+ -rw-r--r--. 1 root root 136M Jun 28 12:49 embeddinggemma-rus-32768-Q8_0.gguf
59
+ -rw-r--r--. 1 root root 8.1K Jun 28 12:49 smoke-embedding.json
60
+
61
+ ## SHA256
62
+ cdc20839de6745e478b5869f5eb5f3275aee79799435446b863f8cd104c79055 embeddinggemma-rus-32768-F32.gguf
63
+ dc7c84e1d37d53344d5d5b617c4db2ebeb928ab147880708e77113062b0f7f14 embeddinggemma-rus-32768-Q8_0.gguf
64
+ a0df320b451b8d513ba8dd74d027c5de7e44dc749db33966fffec0cb87e163b4 smoke-embedding.json
65
+
66
+ conversion_finished_utc=2026-06-28T12:49:52Z
embeddinggemma-rus-32768-F32.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cdc20839de6745e478b5869f5eb5f3275aee79799435446b863f8cd104c79055
3
+ size 527387488
embeddinggemma-rus-32768-Q8_0.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc7c84e1d37d53344d5d5b617c4db2ebeb928ab147880708e77113062b0f7f14
3
+ size 141593440
smoke-embedding.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "object": "list",
3
+ "data": [
4
+ {
5
+ "object": "embedding",
6
+ "index": 0,
7
+ "embedding": [-0.0987985,0.0172395,0.0240678,-0.0207397,0.0230546,0.0037354,0.0224175,-0.0246922,0.0224542,-0.0948961,0.0024555,-0.0683104,0.0161307,-0.0008460,0.0823612,0.0307843,0.0363599,-0.0001738,-0.0636374,0.0599359,0.0011542,0.0724528,0.0659954,-0.0223651,0.0158392,0.0050373,0.0029139,0.0011772,-0.0444994,0.0365572,-0.0603860,0.0192177,0.0028660,-0.0068241,0.0059831,0.0166751,-0.0008945,-0.0241368,0.0092152,0.0202583,-0.0428203,0.0347055,0.0407929,0.0385876,0.0401634,-0.0845434,0.0303002,-0.0154034,-0.0203229,-0.0198710,-0.0711159,0.0197140,0.0050550,-0.0264688,-0.0103648,-0.0170438,-0.0070542,-0.1079820,0.0261631,-0.0588035,-0.0548053,-0.0243648,-0.0550930,-0.0519829,0.0476070,0.0032927,0.0244465,-0.0144839,0.0650341,0.0621148,0.0152530,-0.0134601,-0.0163405,0.0321909,0.1564192,0.0265195,0.0393605,0.0286955,0.0009351,-0.0343172,0.0085441,0.0276107,-0.0496951,-0.0308103,0.0569011,0.0031818,-0.0347525,0.0256860,0.0345134,-0.0314701,0.0688246,-0.0277442,-0.0155146,0.0369830,0.0083563,-0.0518929,-0.0677000,0.0508665,-0.0000482,-0.0020557,-0.0514214,-0.0439162,0.0057653,0.0630711,0.0002145,0.0238158,-0.0462802,-0.0597955,0.0647851,-0.0615820,-0.0174949,0.0029044,-0.0150207,-0.0132151,-0.0103636,-0.0338385,-0.1112098,-0.0109623,-0.0064924,-0.0493083,-0.0301750,-0.0134738,0.0022012,0.0645462,-0.0272656,-0.0134045,-0.0085000,-0.0611257,-0.0638668,-0.0434695,-0.0233153,-0.0160795,0.0036176,-0.0322560,-0.0694082,0.0685000,0.0542858,0.0098594,-0.0301007,-0.0082729,-0.0556583,-0.0015361,0.0143701,-0.0263857,-0.0153483,-0.0312776,-0.0479260,-0.0378133,0.0113807,-0.0314054,0.0132441,0.0406416,0.0377288,-0.0246610,0.0316933,-0.0151471,0.0225853,-0.0369215,0.0709684,-0.0141284,-0.0165644,-0.0468785,-0.0333423,0.0478441,0.0027970,0.0918959,-0.0012772,-0.0078039,0.0095256,-0.0588279,0.0039886,0.0091364,0.0438682,-0.0348059,-0.0519675,0.0145790,0.0081769,-0.0258621,0.0287294,0.0314488,0.0256126,0.0348167,-0.0061517,-0.0337648,-0.0166878,0.0060484,0.0127411,-0.0163152,-0.0319652,0.0011813,0.0075202,0.0240863,-0.0381710,0.0378205,-0.0433501,-0.0459499,0.0582606,0.0103923,0.0014303,-0.0340125,-0.0177122,0.0105075,0.0109808,-0.0008340,-0.0399479,0.0028682,0.0029496,-0.0123172,-0.0601989,-0.0106704,0.0393717,0.0003591,0.0232722,-0.0095811,-0.0441602,0.0197119,-0.0088049,0.0470861,0.0222836,0.0535264,0.0085415,-0.0069235,-0.0003744,-0.0369427,-0.0463668,0.0288451,0.0292538,-0.0302809,0.0370968,-0.0032618,-0.0253300,-0.0140646,0.0368639,-0.0057769,-0.0558192,-0.0927164,-0.0697252,-0.0100701,-0.0374487,0.0606878,-0.0122028,0.0237625,-0.0282304,-0.0283636,0.0788094,0.0004262,0.0469002,0.0220472,0.0209207,-0.0214838,-0.0736573,-0.0102789,0.0469470,0.0238293,-0.0143942,0.0220203,0.0454885,-0.0041058,0.0028326,0.0033147,-0.0031411,0.0239559,-0.0498834,-0.0288431,-0.0362686,0.0566651,0.0511293,0.0056152,-0.0018263,-0.0099692,-0.0338677,-0.0654944,0.0389998,0.0047232,0.0207772,0.0583202,0.0670065,-0.0187493,0.0216162,-0.0115467,-0.0693595,0.0116973,-0.0205904,0.0367825,0.0108164,-0.0321136,-0.0087740,-0.0446232,0.0000194,0.0275580,0.0409758,0.0207485,-0.0710404,0.0002671,-0.0030809,-0.0382422,0.0184755,-0.0540115,-0.0360079,-0.0255372,0.0534095,-0.0472854,0.0162272,0.0172916,0.0399162,0.0209834,0.0417896,0.0294056,-0.0076203,0.0432350,-0.0027797,0.0158175,0.0010203,-0.0220400,0.0705464,0.0853251,-0.0377692,0.0114734,0.0208331,-0.0109571,-0.0256835,-0.0167052,0.0225674,0.0250042,-0.0271485,0.0474790,0.0017990,0.0079934,-0.0687437,-0.0201132,-0.0280700,-0.0286404,0.0696351,-0.0102712,-0.0077082,-0.0197756,-0.0995837,0.0419465,-0.0419453,0.0393926,-0.0149508,-0.0319170,0.0339129,0.0282547,0.0701640,0.0827033,0.0530574,0.0084572,-0.0003019,0.0055859,0.0291131,-0.0267096,-0.0284443,-0.0167570,-0.1195245,-0.0027409,0.0147897,-0.0139943,-0.0039981,0.0122511,0.0787291,-0.0596662,-0.0088841,-0.0604217,0.0182541,0.0051955,-0.0021782,0.0693884,0.0352733,-0.0043835,-0.0475667,-0.0151286,-0.0391483,0.0103732,-0.0162468,-0.0161433,0.0038030,0.0010684,-0.0156499,0.0074679,-0.0261443,0.0150774,0.0439461,-0.0010044,-0.0357697,-0.0049106,-0.0416085,-0.0160580,-0.0319541,0.0066047,-0.0185206,-0.0186645,-0.0112040,0.0143774,-0.0473689,0.0044730,-0.0031880,-0.0399688,-0.0170706,-0.0406776,-0.0243936,-0.0379598,0.0394799,0.0144630,0.0819022,-0.0433157,-0.0298633,-0.0567860,-0.0002802,-0.0180937,0.0037082,-0.0766564,-0.0031718,0.0252293,-0.0090185,-0.0230808,0.0508482,-0.0043524,-0.0341979,0.0217203,-0.0157950,0.0460611,0.0019836,0.0137886,0.0400669,0.0006711,-0.0214250,0.0181733,0.0032269,0.0059862,-0.0293827,-0.0558566,0.0275373,0.0349811,-0.0350947,-0.0143568,-0.0061062,0.0083376,0.0204834,0.0881477,-0.0534020,0.0059177,0.0506220,-0.0282846,0.0483888,0.0049757,0.0302340,-0.0381947,-0.0227967,0.0538965,-0.0044935,0.0577230,0.0005453,0.0261161,0.0716592,-0.0183504,0.0261518,0.0570354,-0.0098746,-0.0104399,0.0532035,-0.0433428,0.0423959,-0.0258443,-0.0349113,0.0072549,0.0350529,-0.0111778,0.0028155,0.0482814,-0.0042423,0.0439919,0.0380178,-0.0644180,-0.0041264,-0.0081390,0.0134051,0.0278067,-0.0625829,-0.0667183,0.0161716,0.0141715,-0.0058942,-0.0663553,-0.0115408,-0.0562086,0.0267148,-0.0734466,-0.0161264,0.0204283,0.0349491,-0.0113434,0.0110507,0.0142087,0.0369580,-0.0037961,0.0416785,-0.0322044,0.0221956,-0.0454602,0.0030769,0.0252344,0.0451122,0.0105065,-0.0388315,0.0312859,0.0521952,-0.0373766,0.0684618,0.0058506,-0.0432100,-0.0202228,0.0131115,0.0122266,-0.0235783,-0.0202008,0.0031743,-0.0057063,-0.0271225,-0.0015862,0.0193878,0.0280173,0.0399563,-0.0095105,0.0059835,0.0285771,0.0239451,0.0207253,0.0245928,-0.0433706,-0.0254622,-0.0428807,0.0497240,-0.0044434,0.0185679,0.0287334,0.0215057,-0.0463180,-0.0966902,-0.0164682,-0.0616246,-0.0425205,-0.0286170,-0.0386116,0.0139175,-0.0066561,-0.0234127,-0.0458217,-0.0033630,-0.0090089,-0.0240107,0.0420749,-0.0454043,-0.0191317,0.0058141,0.0342979,0.0048587,-0.0338035,0.0293148,0.0173548,0.0020753,-0.0147408,-0.0096575,-0.0249928,0.0552181,-0.0170997,0.0305983,0.0116961,0.0195360,-0.0276793,-0.0324858,-0.0140775,-0.0535239,-0.0236582,0.0111463,0.0042716,-0.0511280,-0.0553730,-0.0285782,0.0267802,0.0244693,-0.0326682,0.0222497,-0.0456699,0.0303788,-0.0553025,-0.0781417,0.0416733,-0.0146209,0.0242698,0.0429391,-0.0108855,-0.0173534,-0.0061078,-0.0664671,0.0349651,0.0407880,0.0099391,0.0269564,0.0140259,-0.0180590,0.0224159,-0.0415195,-0.0578230,-0.0016462,-0.0067796,0.0045791,0.0359983,-0.0266611,0.0481297,-0.0465197,-0.0065630,-0.0420411,0.0050263,-0.0333450,-0.0190896,-0.0246505,-0.0182685,-0.0126475,-0.0021452,0.0724644,-0.0137872,-0.0025639,-0.0296527,-0.0324376,-0.0090404,0.0493689,-0.0529604,-0.0458398,-0.0650851,0.0057058,0.0340488,0.0721952,0.0218946,-0.0093925,-0.0043259,-0.0761353,0.0144927,0.0097741,-0.0121322,-0.0023223,-0.0086623,0.0187595,0.0272376,-0.0363487,-0.0493621,0.0386689,-0.0330862,-0.0179095,-0.0020520,0.0136226,0.0123670,0.0344815,-0.0215708,0.0554766,0.0189449,-0.0657016,0.0355155,0.0519469,0.0140132,0.0112173,-0.0107099,0.0172485,0.0081740,-0.0393141,-0.0085314,0.0378849,-0.0081039,-0.0183365,0.0063688,-0.0628638,0.0048059,0.0347058,-0.0189910,0.0283956,0.0426782,-0.0121495,-0.0645409,-0.0361222,0.0405916,-0.0079862,-0.0178930,-0.0128091,0.0212762,0.0112469,0.0148284,-0.0005971,-0.0570026,0.0186069,0.0037429,0.0004663,0.0168569,0.0271739,-0.0207333,0.0086891,-0.0173462,0.0268942,0.0054576,0.0344776,-0.0401311,0.0771179,-0.0169458,-0.0544396,-0.0296540,0.0331283,-0.0055856,0.0000412,-0.0220357,0.0070481,0.0909272,-0.0358025,-0.0050808,-0.0507158,-0.0198469,-0.0055818,0.0311455,-0.0553912,-0.0916804,-0.0332116,-0.0460665,-0.0075208,-0.0303838,0.0274890,-0.0361109,-0.0048877,0.0579376,0.0429564,0.0030423,0.0129577,0.0011942,-0.0201044,-0.0136231,0.0131161,0.0062296,-0.0087274,-0.0485708,0.0010869,-0.0749795,-0.0165168,0.0142147,-0.0005397,0.0052404,-0.0411452,0.0273822,-0.0102311,-0.0075292,-0.0221791,0.0277793,0.0194753,0.0591136,-0.0428330,-0.0098394,0.0234091,0.0126999,0.0062244,-0.0499975,-0.0488437,-0.0030458,-0.0567850,0.0239679,0.0659956,0.0092680,-0.0125627,-0.0149546,0.0382171,-0.0007062,-0.0226019,0.0298504]
8
+ }
9
+ ]
10
+ }
11
+