Hugging Face's logo

tss-deposium
/
harrier-oss-v1-270m-onnx-int8

Feature Extraction
Transformers.js
ONNX
multilingual
gemma3_text
embedding
int8
harrier
Model card Files
xet
 Community
tss-deposium commited on
Commit
b889c60
·
verified ·
1 Parent(s): 5eb6942

harrier-oss-v1-270m INT8 single-file — 270MB, 640D, 94 langs

Browse files
Files changed (8) hide show
  1. .gitattributes +1 -0
  2. README.md +26 -0
  3. config.json +58 -0
  4. onnx/model.onnx +3 -0
  5. search_instructions.json +8 -0
  6. special_tokens_map.json +33 -0
  7. tokenizer.json +3 -0
  8. tokenizer_config.json +0 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* 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
 
 
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
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: microsoft/harrier-oss-v1-270m
3
+ library_name: transformers.js
4
+ license: mit
5
+ language: [multilingual]
6
+ tags: [embedding, onnx, int8, transformers.js, harrier, multilingual]
7
+ pipeline_tag: feature-extraction
8
+ ---
9
+ # Harrier-OSS-v1-270M — ONNX INT8 (single-file)
10
+
11
+ Re-export of [microsoft/harrier-oss-v1-270m](https://huggingface.co/microsoft/harrier-oss-v1-270m)
12
+ as single-file ONNX INT8 for Transformers.js v4 and ONNX Runtime.
13
+
14
+ | Property | Value |
15
+ |----------|-------|
16
+ | Dimensions | 640 |
17
+ | MTEB v2 | 66.5 |
18
+ | Languages | 94 |
19
+ | Max tokens | 32,768 |
20
+ | Pooling | Last-token + L2 norm |
21
+ | Size | 270 MB (INT8) |
22
+ | License | MIT |
23
+
24
+ **Important**: Queries must be prefixed with an instruction. See `search_instructions.json`.
25
+
26
+ Export by [Deposium](https://deposium.ai).
config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_sliding_window_pattern": 1,
3
+ "architectures": [
4
+ "Gemma3TextModel"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_logit_softcapping": null,
9
+ "bos_token_id": 2,
10
+ "dtype": "bfloat16",
11
+ "eos_token_id": 1,
12
+ "final_logit_softcapping": null,
13
+ "head_dim": 256,
14
+ "hidden_activation": "gelu_pytorch_tanh",
15
+ "hidden_size": 640,
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 2048,
18
+ "layer_types": [
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention"
37
+ ],
38
+ "max_position_embeddings": 32768,
39
+ "model_type": "gemma3_text",
40
+ "num_attention_heads": 4,
41
+ "num_hidden_layers": 18,
42
+ "num_key_value_heads": 1,
43
+ "pad_token_id": 0,
44
+ "query_pre_attn_scalar": 256,
45
+ "rms_norm_eps": 1e-06,
46
+ "rope_local_base_freq": 10000.0,
47
+ "rope_scaling": null,
48
+ "rope_theta": 1000000.0,
49
+ "sliding_window": 512,
50
+ "sliding_window_pattern": 1,
51
+ "transformers_version": "4.57.6",
52
+ "use_bidirectional_attention": false,
53
+ "use_cache": false,
54
+ "vocab_size": 262144,
55
+ "transformers.js_config": {
56
+ "model_file_name": "model"
57
+ }
58
+ }
onnx/model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:08fd8f7fdd0769289d012f89d5ae82452ba1fa95d394358615f4280a45a0f460
3
+ size 270135343
search_instructions.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "query_prefix": "Instruct: Given a web search query, retrieve relevant passages that answer the query\nQuery: ",
3
+ "document_prefix": "",
4
+ "note": "Queries MUST be prefixed with query_prefix. Documents must NOT be prefixed.",
5
+ "pooling": "last_token",
6
+ "normalize": "l2",
7
+ "dimensions": 640
8
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "boi_token": "<start_of_image>",
3
+ "bos_token": {
4
+ "content": "<bos>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "eoi_token": "<end_of_image>",
11
+ "eos_token": {
12
+ "content": "<eos>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false
17
+ },
18
+ "image_token": "<image_soft_token>",
19
+ "pad_token": {
20
+ "content": "<pad>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false
25
+ },
26
+ "unk_token": {
27
+ "content": "<unk>",
28
+ "lstrip": false,
29
+ "normalized": false,
30
+ "rstrip": false,
31
+ "single_word": false
32
+ }
33
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6852f8d561078cc0cebe70ca03c5bfdd0d60a45f9d2e0e1e4cc05b68e9ec329e
3
+ size 33385008
tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff