bwang0911 commited on
Commit
3301f21
·
verified ·
1 Parent(s): 1489172

Upload 4 files

Browse files
Files changed (1) hide show
  1. config.json +7 -14
config.json CHANGED
@@ -1,12 +1,12 @@
1
  {
2
- "_name_or_path": "bwang0911/test-jina-xlm-roberta-ft",
3
  "architectures": [
4
  "XLMRobertaModel"
5
  ],
6
  "attention_probs_dropout_prob": 0.1,
7
  "auto_map": {
8
  "AutoConfig": "jinaai/xlm-roberta-flash-implementation--configuration_xlm_roberta.XLMRobertaFlashConfig",
9
- "AutoModel": "jinaai/xlm-roberta-flash-implementation--modeling_lora.XLMRobertaLoRA",
10
  "AutoModelForMaskedLM": "jinaai/xlm-roberta-flash-implementation--modeling_xlm_roberta.XLMRobertaForMaskedLM",
11
  "AutoModelForPreTraining": "jinaai/xlm-roberta-flash-implementation--modeling_xlm_roberta.XLMRobertaForPreTraining"
12
  },
@@ -20,32 +20,25 @@
20
  "initializer_range": 0.02,
21
  "intermediate_size": 4096,
22
  "layer_norm_eps": 1e-05,
23
- "load_trained_adapters": true,
24
- "lora_adaptations": ["retrieval.query", "retrieval.passage", "separation", "classification", "text-matching"],
25
  "lora_alpha": 1,
26
  "lora_dropout_p": 0.0,
27
  "lora_main_params_trainable": false,
28
  "lora_rank": 4,
29
- "matryoshka_dimensions": [32, 64, 128, 256, 512, 768, 1024],
30
  "max_position_embeddings": 8194,
31
  "num_attention_heads": 16,
32
  "num_hidden_layers": 24,
33
  "output_past": true,
34
  "pad_token_id": 1,
35
  "position_embedding_type": "rotary",
36
- "rotary_emb_base": 20000.0,
37
  "torch_dtype": "bfloat16",
38
  "transformers_version": "4.30.2",
39
  "truncate_dim": null,
40
  "type_vocab_size": 1,
41
  "use_cache": true,
42
  "use_flash_attn": true,
43
- "vocab_size": 250002,
44
- "task_instructions": {
45
- "retrieval.query": "Represent the query for retrieving evidence documents: ",
46
- "retrieval.passage": "Represent the document for retrieval: ",
47
- "separation": "",
48
- "classification": "",
49
- "text-matching": ""
50
- }
51
  }
 
1
  {
2
+ "_name_or_path": "jinaai/jina-embeddings-v3-xlm-large-pairs-vol3-temp-0.02",
3
  "architectures": [
4
  "XLMRobertaModel"
5
  ],
6
  "attention_probs_dropout_prob": 0.1,
7
  "auto_map": {
8
  "AutoConfig": "jinaai/xlm-roberta-flash-implementation--configuration_xlm_roberta.XLMRobertaFlashConfig",
9
+ "AutoModel": "jinaai/xlm-roberta-flash-implementation--modeling_xlm_roberta.XLMRobertaModel",
10
  "AutoModelForMaskedLM": "jinaai/xlm-roberta-flash-implementation--modeling_xlm_roberta.XLMRobertaForMaskedLM",
11
  "AutoModelForPreTraining": "jinaai/xlm-roberta-flash-implementation--modeling_xlm_roberta.XLMRobertaForPreTraining"
12
  },
 
20
  "initializer_range": 0.02,
21
  "intermediate_size": 4096,
22
  "layer_norm_eps": 1e-05,
23
+ "load_trained_adapters": false,
24
+ "lora_adaptations": null,
25
  "lora_alpha": 1,
26
  "lora_dropout_p": 0.0,
27
  "lora_main_params_trainable": false,
28
  "lora_rank": 4,
29
+ "matryoshka_dimensions": null,
30
  "max_position_embeddings": 8194,
31
  "num_attention_heads": 16,
32
  "num_hidden_layers": 24,
33
  "output_past": true,
34
  "pad_token_id": 1,
35
  "position_embedding_type": "rotary",
36
+ "rotary_emb_base": 10000.0,
37
  "torch_dtype": "bfloat16",
38
  "transformers_version": "4.30.2",
39
  "truncate_dim": null,
40
  "type_vocab_size": 1,
41
  "use_cache": true,
42
  "use_flash_attn": true,
43
+ "vocab_size": 250002
 
 
 
 
 
 
 
44
  }