Hugging Face's logo

genshiai-daichi
/
med-ruri-v3-70m-v2-from-ruri

Sentence Similarity
sentence-transformers
Safetensors
Japanese
modernbert
medical
japanese
ruri
embedding
text-embeddings-inference
Model card Files
xet
 Community
genshiai-daichi commited on
Commit
a943ace
·
verified ·
1 Parent(s): 639f443

latest = step 500 (nDCG@10=0.3859)

Browse files
Files changed (9) hide show
  1. 1_Pooling/config.json +5 -0
  2. config.json +70 -0
  3. config_sentence_transformers.json +15 -0
  4. model.safetensors +3 -0
  5. modules.json +14 -0
  6. sentence_bert_config.json +10 -0
  7. tokenizer.json +0 -0
  8. tokenizer_config.json +25 -0
  9. trainer_state.json +123 -0
1_Pooling/config.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "embedding_dimension": 384,
3
+ "pooling_mode": "mean",
4
+ "include_prompt": true
5
+ }
config.json ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ModernBertModel"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 1,
8
+ "classifier_activation": "gelu",
9
+ "classifier_bias": false,
10
+ "classifier_dropout": 0.0,
11
+ "classifier_pooling": "cls",
12
+ "cls_token_id": 6,
13
+ "decoder_bias": true,
14
+ "deterministic_flash_attn": false,
15
+ "dtype": "float32",
16
+ "embedding_dropout": 0.0,
17
+ "eos_token_id": 2,
18
+ "global_attn_every_n_layers": 3,
19
+ "gradient_checkpointing": false,
20
+ "hidden_activation": "gelu",
21
+ "hidden_size": 384,
22
+ "initializer_cutoff_factor": 2.0,
23
+ "initializer_range": 0.02,
24
+ "intermediate_size": 1536,
25
+ "layer_norm_eps": 1e-05,
26
+ "layer_types": [
27
+ "full_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "full_attention",
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "full_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "full_attention",
37
+ "sliding_attention",
38
+ "sliding_attention",
39
+ "full_attention"
40
+ ],
41
+ "local_attention": 128,
42
+ "max_position_embeddings": 8192,
43
+ "mlp_bias": false,
44
+ "mlp_dropout": 0.0,
45
+ "model_type": "modernbert",
46
+ "norm_bias": false,
47
+ "norm_eps": 1e-05,
48
+ "num_attention_heads": 6,
49
+ "num_hidden_layers": 13,
50
+ "pad_token_id": 3,
51
+ "position_embedding_type": "rope",
52
+ "repad_logits_with_grad": false,
53
+ "rope_parameters": {
54
+ "full_attention": {
55
+ "rope_theta": 160000.0,
56
+ "rope_type": "default"
57
+ },
58
+ "sliding_attention": {
59
+ "rope_theta": 10000.0,
60
+ "rope_type": "default"
61
+ }
62
+ },
63
+ "sep_token_id": 4,
64
+ "sparse_pred_ignore_index": -100,
65
+ "sparse_prediction": false,
66
+ "tie_word_embeddings": true,
67
+ "transformers_version": "5.8.1",
68
+ "use_cache": false,
69
+ "vocab_size": 102400
70
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "pytorch": "2.11.0+cu128",
4
+ "sentence_transformers": "5.5.1",
5
+ "transformers": "5.8.1"
6
+ },
7
+ "default_prompt_name": null,
8
+ "model_type": "SentenceTransformer",
9
+ "prompts": {
10
+ "document": "\u691c\u7d22\u6587\u66f8: ",
11
+ "query": "\u691c\u7d22\u30af\u30a8\u30ea: ",
12
+ "topic": "\u30c8\u30d4\u30c3\u30af: "
13
+ },
14
+ "similarity_fn_name": "cosine"
15
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88bbef91becf4ba2e457ba35c9dabe61b15e0c85435c9526ab753c7aa689e6b8
3
+ size 280019040
modules.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "idx": 0,
4
+ "name": "0",
5
+ "path": "",
6
+ "type": "sentence_transformers.base.modules.transformer.Transformer"
7
+ },
8
+ {
9
+ "idx": 1,
10
+ "name": "1",
11
+ "path": "1_Pooling",
12
+ "type": "sentence_transformers.sentence_transformer.modules.pooling.Pooling"
13
+ }
14
+ ]
sentence_bert_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "transformer_task": "feature-extraction",
3
+ "modality_config": {
4
+ "text": {
5
+ "method": "forward",
6
+ "method_output_name": "last_hidden_state"
7
+ }
8
+ },
9
+ "module_output_name": "token_embeddings"
10
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_dummy_prefix_space": false,
3
+ "add_prefix_space": false,
4
+ "backend": "tokenizers",
5
+ "bos_token": "<s>",
6
+ "clean_up_tokenization_spaces": false,
7
+ "cls_token": "<cls>",
8
+ "do_lower_case": false,
9
+ "eos_token": "</s>",
10
+ "extra_ids": 0,
11
+ "is_local": false,
12
+ "keep_accents": true,
13
+ "legacy": false,
14
+ "local_files_only": false,
15
+ "mask_token": "<mask>",
16
+ "model_max_length": 768,
17
+ "pad_token": "<pad>",
18
+ "padding_side": "right",
19
+ "sep_token": "<sep>",
20
+ "sp_model_kwargs": {},
21
+ "spaces_between_special_tokens": false,
22
+ "tokenizer_class": "TokenizersBackend",
23
+ "unk_token": "<unk>",
24
+ "use_default_system_prompt": false
25
+ }
trainer_state.json ADDED
@@ -0,0 +1,123 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 500,
3
+ "best_metric": 0.3859062398415415,
4
+ "best_model_checkpoint": "checkpoints/med-ruri-v3-70m-v2-from-ruri/checkpoint-500",
5
+ "epoch": 0.08118201006656925,
6
+ "eval_steps": 500,
7
+ "global_step": 500,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.008118201006656925,
14
+ "grad_norm": 21.997957229614258,
15
+ "learning_rate": 1.590909090909091e-06,
16
+ "loss": 2.7372756958007813,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.01623640201331385,
21
+ "grad_norm": 21.73430824279785,
22
+ "learning_rate": 3.2142857142857147e-06,
23
+ "loss": 1.9263325500488282,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.024354603019970774,
28
+ "grad_norm": 20.49755859375,
29
+ "learning_rate": 4.837662337662339e-06,
30
+ "loss": 1.7200509643554687,
31
+ "step": 150
32
+ },
33
+ {
34
+ "epoch": 0.0324728040266277,
35
+ "grad_norm": 21.036407470703125,
36
+ "learning_rate": 6.461038961038961e-06,
37
+ "loss": 1.5285859680175782,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 0.040591005033284625,
42
+ "grad_norm": 21.67646026611328,
43
+ "learning_rate": 8.084415584415586e-06,
44
+ "loss": 1.4216513061523437,
45
+ "step": 250
46
+ },
47
+ {
48
+ "epoch": 0.04870920603994155,
49
+ "grad_norm": 30.95376205444336,
50
+ "learning_rate": 9.707792207792209e-06,
51
+ "loss": 1.382916259765625,
52
+ "step": 300
53
+ },
54
+ {
55
+ "epoch": 0.05682740704659847,
56
+ "grad_norm": 22.470050811767578,
57
+ "learning_rate": 9.929926508289183e-06,
58
+ "loss": 1.316688995361328,
59
+ "step": 350
60
+ },
61
+ {
62
+ "epoch": 0.0649456080532554,
63
+ "grad_norm": 31.66095733642578,
64
+ "learning_rate": 9.84447103059306e-06,
65
+ "loss": 1.284674072265625,
66
+ "step": 400
67
+ },
68
+ {
69
+ "epoch": 0.07306380905991232,
70
+ "grad_norm": 20.316326141357422,
71
+ "learning_rate": 9.759015552896942e-06,
72
+ "loss": 1.18936767578125,
73
+ "step": 450
74
+ },
75
+ {
76
+ "epoch": 0.08118201006656925,
77
+ "grad_norm": 23.299427032470703,
78
+ "learning_rate": 9.67356007520082e-06,
79
+ "loss": 1.1284925842285156,
80
+ "step": 500
81
+ },
82
+ {
83
+ "epoch": 0.08118201006656925,
84
+ "eval_med-ir_cosine_accuracy@1": 0.2215,
85
+ "eval_med-ir_cosine_accuracy@10": 0.5745,
86
+ "eval_med-ir_cosine_accuracy@5": 0.4715,
87
+ "eval_med-ir_cosine_map@10": 0.32694821428571424,
88
+ "eval_med-ir_cosine_mrr@10": 0.3269482142857145,
89
+ "eval_med-ir_cosine_ndcg@10": 0.3859062398415415,
90
+ "eval_med-ir_cosine_precision@1": 0.2215,
91
+ "eval_med-ir_cosine_precision@10": 0.057449999999999994,
92
+ "eval_med-ir_cosine_precision@5": 0.0943,
93
+ "eval_med-ir_cosine_recall@1": 0.2215,
94
+ "eval_med-ir_cosine_recall@10": 0.5745,
95
+ "eval_med-ir_cosine_recall@5": 0.4715,
96
+ "eval_runtime": 53.6164,
97
+ "eval_samples_per_second": 0.0,
98
+ "eval_steps_per_second": 0.0,
99
+ "step": 500
100
+ }
101
+ ],
102
+ "logging_steps": 50,
103
+ "max_steps": 6159,
104
+ "num_input_tokens_seen": 0,
105
+ "num_train_epochs": 1,
106
+ "save_steps": 500,
107
+ "stateful_callbacks": {
108
+ "TrainerControl": {
109
+ "args": {
110
+ "should_epoch_stop": false,
111
+ "should_evaluate": false,
112
+ "should_log": false,
113
+ "should_save": true,
114
+ "should_training_stop": false
115
+ },
116
+ "attributes": {}
117
+ }
118
+ },
119
+ "total_flos": 0.0,
120
+ "train_batch_size": 192,
121
+ "trial_name": null,
122
+ "trial_params": null
123
+ }