Upload optimized model
Browse files- README.md +15 -0
- config.json +52 -0
- model.safetensors +3 -0
README.md
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- quantized
|
| 4 |
+
- embedding_finetuning
|
| 5 |
+
- 4-bit
|
| 6 |
+
library_name: optimum
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
# Model - tempe-quantized
|
| 10 |
+
|
| 11 |
+
This model has been optimized and uploaded to the HuggingFace Hub.
|
| 12 |
+
|
| 13 |
+
## Model Details
|
| 14 |
+
- Original Repository: tempe-quantized
|
| 15 |
+
- Optimization Tags: quantized, embedding_finetuning, 4-bit
|
config.json
ADDED
|
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "jinaai/jina-embeddings-v2-base-zh",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertModel"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"auto_map": {
|
| 8 |
+
"AutoConfig": "jinaai/jina-bert-implementation--configuration_bert.JinaBertConfig",
|
| 9 |
+
"AutoModel": "jinaai/jina-bert-implementation--modeling_bert.JinaBertModel",
|
| 10 |
+
"AutoModelForMaskedLM": "jinaai/jina-bert-implementation--modeling_bert.JinaBertForMaskedLM",
|
| 11 |
+
"AutoModelForQuestionAnswering": "jinaai/jina-bert-implementation--modeling_bert.JinaBertForQuestionAnswering",
|
| 12 |
+
"AutoModelForSequenceClassification": "jinaai/jina-bert-implementation--modeling_bert.JinaBertForSequenceClassification",
|
| 13 |
+
"AutoModelForTokenClassification": "jinaai/jina-bert-implementation--modeling_bert.JinaBertForTokenClassification"
|
| 14 |
+
},
|
| 15 |
+
"classifier_dropout": null,
|
| 16 |
+
"emb_pooler": "mean",
|
| 17 |
+
"feed_forward_type": "geglu",
|
| 18 |
+
"gradient_checkpointing": false,
|
| 19 |
+
"hidden_act": "gelu",
|
| 20 |
+
"hidden_dropout_prob": 0.1,
|
| 21 |
+
"hidden_size": 768,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 3072,
|
| 24 |
+
"layer_norm_eps": 1e-12,
|
| 25 |
+
"max_position_embeddings": 8192,
|
| 26 |
+
"model_max_length": 8192,
|
| 27 |
+
"model_type": "bert",
|
| 28 |
+
"num_attention_heads": 12,
|
| 29 |
+
"num_hidden_layers": 12,
|
| 30 |
+
"pad_token_id": 0,
|
| 31 |
+
"position_embedding_type": "alibi",
|
| 32 |
+
"quantization_config": {
|
| 33 |
+
"_load_in_4bit": true,
|
| 34 |
+
"_load_in_8bit": false,
|
| 35 |
+
"bnb_4bit_compute_dtype": "float32",
|
| 36 |
+
"bnb_4bit_quant_storage": "uint8",
|
| 37 |
+
"bnb_4bit_quant_type": "fp4",
|
| 38 |
+
"bnb_4bit_use_double_quant": false,
|
| 39 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
| 40 |
+
"llm_int8_has_fp16_weight": false,
|
| 41 |
+
"llm_int8_skip_modules": null,
|
| 42 |
+
"llm_int8_threshold": 6.0,
|
| 43 |
+
"load_in_4bit": true,
|
| 44 |
+
"load_in_8bit": false,
|
| 45 |
+
"quant_method": "bitsandbytes"
|
| 46 |
+
},
|
| 47 |
+
"torch_dtype": "float16",
|
| 48 |
+
"transformers_version": "4.47.1",
|
| 49 |
+
"type_vocab_size": 2,
|
| 50 |
+
"use_cache": true,
|
| 51 |
+
"vocab_size": 61056
|
| 52 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:08dd54c62678767e447bc392fa09e00ed49765480f8b0b707559b6fc864919bb
|
| 3 |
+
size 153019119
|