Upload folder using huggingface_hub
Browse files- .gitattributes +6 -0
- 1_Pooling/config.json +10 -0
- README.md +17 -0
- checkpoint-1564/1_Pooling/config.json +10 -0
- checkpoint-1564/README.md +542 -0
- checkpoint-1564/config.json +26 -0
- checkpoint-1564/config_sentence_transformers.json +10 -0
- checkpoint-1564/model.safetensors +3 -0
- checkpoint-1564/modules.json +14 -0
- checkpoint-1564/optimizer.pt +3 -0
- checkpoint-1564/rng_state.pth +3 -0
- checkpoint-1564/scheduler.pt +3 -0
- checkpoint-1564/sentence_bert_config.json +4 -0
- checkpoint-1564/special_tokens_map.json +51 -0
- checkpoint-1564/tokenizer.json +3 -0
- checkpoint-1564/tokenizer_config.json +64 -0
- checkpoint-1564/trainer_state.json +512 -0
- checkpoint-1564/training_args.bin +3 -0
- checkpoint-1564/unigram.json +3 -0
- checkpoint-3128/1_Pooling/config.json +10 -0
- checkpoint-3128/README.md +610 -0
- checkpoint-3128/config.json +26 -0
- checkpoint-3128/config_sentence_transformers.json +10 -0
- checkpoint-3128/model.safetensors +3 -0
- checkpoint-3128/modules.json +14 -0
- checkpoint-3128/optimizer.pt +3 -0
- checkpoint-3128/rng_state.pth +3 -0
- checkpoint-3128/scheduler.pt +3 -0
- checkpoint-3128/sentence_bert_config.json +4 -0
- checkpoint-3128/special_tokens_map.json +51 -0
- checkpoint-3128/tokenizer.json +3 -0
- checkpoint-3128/tokenizer_config.json +64 -0
- checkpoint-3128/trainer_state.json +989 -0
- checkpoint-3128/training_args.bin +3 -0
- checkpoint-3128/unigram.json +3 -0
- config.json +26 -0
- config_sentence_transformers.json +10 -0
- model.safetensors +3 -0
- modules.json +14 -0
- runs/May29_18-52-15_338a77628651/events.out.tfevents.1717008737.338a77628651.19835.0 +2 -2
- sentence_bert_config.json +4 -0
- special_tokens_map.json +51 -0
- tokenizer.json +3 -0
- tokenizer_config.json +64 -0
- training_args.bin +3 -0
- training_params.json +33 -0
- unigram.json +3 -0
.gitattributes
CHANGED
@@ -33,3 +33,9 @@ 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 |
+
checkpoint-1564/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
37 |
+
checkpoint-1564/unigram.json filter=lfs diff=lfs merge=lfs -text
|
38 |
+
checkpoint-3128/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
39 |
+
checkpoint-3128/unigram.json filter=lfs diff=lfs merge=lfs -text
|
40 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
41 |
+
unigram.json filter=lfs diff=lfs merge=lfs -text
|
1_Pooling/config.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"word_embedding_dimension": 384,
|
3 |
+
"pooling_mode_cls_token": false,
|
4 |
+
"pooling_mode_mean_tokens": true,
|
5 |
+
"pooling_mode_max_tokens": false,
|
6 |
+
"pooling_mode_mean_sqrt_len_tokens": false,
|
7 |
+
"pooling_mode_weightedmean_tokens": false,
|
8 |
+
"pooling_mode_lasttoken": false,
|
9 |
+
"include_prompt": true
|
10 |
+
}
|
README.md
ADDED
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
|
2 |
+
---
|
3 |
+
tags:
|
4 |
+
- autotrain
|
5 |
+
- sentence-transformers
|
6 |
+
widget:
|
7 |
+
- text: "I love AutoTrain"
|
8 |
+
datasets:
|
9 |
+
- ucsahin/TR-Extractive-QA-5K
|
10 |
+
---
|
11 |
+
|
12 |
+
# Model Trained Using AutoTrain
|
13 |
+
|
14 |
+
- Problem type: Sentence Transformers
|
15 |
+
|
16 |
+
## Validation Metrics
|
17 |
+
No validation metrics available
|
checkpoint-1564/1_Pooling/config.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"word_embedding_dimension": 384,
|
3 |
+
"pooling_mode_cls_token": false,
|
4 |
+
"pooling_mode_mean_tokens": true,
|
5 |
+
"pooling_mode_max_tokens": false,
|
6 |
+
"pooling_mode_mean_sqrt_len_tokens": false,
|
7 |
+
"pooling_mode_weightedmean_tokens": false,
|
8 |
+
"pooling_mode_lasttoken": false,
|
9 |
+
"include_prompt": true
|
10 |
+
}
|
checkpoint-1564/README.md
ADDED
@@ -0,0 +1,542 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- multilingual
|
4 |
+
- zh
|
5 |
+
- ja
|
6 |
+
- ar
|
7 |
+
- ko
|
8 |
+
- de
|
9 |
+
- fr
|
10 |
+
- es
|
11 |
+
- pt
|
12 |
+
- hi
|
13 |
+
- id
|
14 |
+
- it
|
15 |
+
- tr
|
16 |
+
- ru
|
17 |
+
- bn
|
18 |
+
- ur
|
19 |
+
- mr
|
20 |
+
- ta
|
21 |
+
- vi
|
22 |
+
- fa
|
23 |
+
- pl
|
24 |
+
- uk
|
25 |
+
- nl
|
26 |
+
- sv
|
27 |
+
- he
|
28 |
+
- sw
|
29 |
+
- ps
|
30 |
+
library_name: sentence-transformers
|
31 |
+
tags:
|
32 |
+
- sentence-transformers
|
33 |
+
- sentence-similarity
|
34 |
+
- feature-extraction
|
35 |
+
- dataset_size:10K<n<100K
|
36 |
+
- loss:CoSENTLoss
|
37 |
+
base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
|
38 |
+
metrics:
|
39 |
+
- pearson_cosine
|
40 |
+
- spearman_cosine
|
41 |
+
- pearson_manhattan
|
42 |
+
- spearman_manhattan
|
43 |
+
- pearson_euclidean
|
44 |
+
- spearman_euclidean
|
45 |
+
- pearson_dot
|
46 |
+
- spearman_dot
|
47 |
+
- pearson_max
|
48 |
+
- spearman_max
|
49 |
+
widget:
|
50 |
+
- source_sentence: Is that wrong?
|
51 |
+
sentences:
|
52 |
+
- Is that such a terrible thing?
|
53 |
+
- Kennedy korkunç bir savcıydı.
|
54 |
+
- Tom bir davada tanıklık ediyordu.
|
55 |
+
- source_sentence: Orada mıydılar?
|
56 |
+
sentences:
|
57 |
+
- Were they in there?
|
58 |
+
- İlki ikincisini anlamlı kılar.
|
59 |
+
- Alerji tedavisi gelişiyor.
|
60 |
+
- source_sentence: He is not alone
|
61 |
+
sentences:
|
62 |
+
- It is not confusing
|
63 |
+
- The Hawks were humanitarians.
|
64 |
+
- Tom bir davada tanıklık ediyordu.
|
65 |
+
- source_sentence: Yaptığın şey bu.
|
66 |
+
sentences:
|
67 |
+
- Onurlu işler yapıyorsunuz.
|
68 |
+
- Weisberg azınlık adına konuştu.
|
69 |
+
- Robert Ferrigno Kaliforniya'da doğdu.
|
70 |
+
- source_sentence: Ben vatansızım.
|
71 |
+
sentences:
|
72 |
+
- I am stateless.
|
73 |
+
- Kendi tekniğini tercih ediyor.
|
74 |
+
- Mermiler camdan fırladı.
|
75 |
+
pipeline_tag: sentence-similarity
|
76 |
+
model-index:
|
77 |
+
- name: SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
|
78 |
+
results:
|
79 |
+
- task:
|
80 |
+
type: semantic-similarity
|
81 |
+
name: Semantic Similarity
|
82 |
+
dataset:
|
83 |
+
name: tr ling
|
84 |
+
type: tr_ling
|
85 |
+
metrics:
|
86 |
+
- type: pearson_cosine
|
87 |
+
value: 0.037604255015168134
|
88 |
+
name: Pearson Cosine
|
89 |
+
- type: spearman_cosine
|
90 |
+
value: 0.04804112988506346
|
91 |
+
name: Spearman Cosine
|
92 |
+
- type: pearson_manhattan
|
93 |
+
value: 0.034740275152181296
|
94 |
+
name: Pearson Manhattan
|
95 |
+
- type: spearman_manhattan
|
96 |
+
value: 0.03769766156967754
|
97 |
+
name: Spearman Manhattan
|
98 |
+
- type: pearson_euclidean
|
99 |
+
value: 0.03698411306484619
|
100 |
+
name: Pearson Euclidean
|
101 |
+
- type: spearman_euclidean
|
102 |
+
value: 0.03903062430281842
|
103 |
+
name: Spearman Euclidean
|
104 |
+
- type: pearson_dot
|
105 |
+
value: 0.0673696846368413
|
106 |
+
name: Pearson Dot
|
107 |
+
- type: spearman_dot
|
108 |
+
value: 0.06818119362900125
|
109 |
+
name: Spearman Dot
|
110 |
+
- type: pearson_max
|
111 |
+
value: 0.0673696846368413
|
112 |
+
name: Pearson Max
|
113 |
+
- type: spearman_max
|
114 |
+
value: 0.06818119362900125
|
115 |
+
name: Spearman Max
|
116 |
+
---
|
117 |
+
|
118 |
+
# SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
|
119 |
+
|
120 |
+
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2) on the [MoritzLaurer/multilingual-nli-26lang-2mil7](https://huggingface.co/datasets/MoritzLaurer/multilingual-nli-26lang-2mil7) dataset. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
|
121 |
+
|
122 |
+
## Model Details
|
123 |
+
|
124 |
+
### Model Description
|
125 |
+
- **Model Type:** Sentence Transformer
|
126 |
+
- **Base model:** [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2) <!-- at revision bf3bf13ab40c3157080a7ab344c831b9ad18b5eb -->
|
127 |
+
- **Maximum Sequence Length:** 128 tokens
|
128 |
+
- **Output Dimensionality:** 384 tokens
|
129 |
+
- **Similarity Function:** Cosine Similarity
|
130 |
+
- **Training Dataset:**
|
131 |
+
- [MoritzLaurer/multilingual-nli-26lang-2mil7](https://huggingface.co/datasets/MoritzLaurer/multilingual-nli-26lang-2mil7)
|
132 |
+
- **Languages:** multilingual, zh, ja, ar, ko, de, fr, es, pt, hi, id, it, tr, ru, bn, ur, mr, ta, vi, fa, pl, uk, nl, sv, he, sw, ps
|
133 |
+
<!-- - **License:** Unknown -->
|
134 |
+
|
135 |
+
### Model Sources
|
136 |
+
|
137 |
+
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
|
138 |
+
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
|
139 |
+
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
|
140 |
+
|
141 |
+
### Full Model Architecture
|
142 |
+
|
143 |
+
```
|
144 |
+
SentenceTransformer(
|
145 |
+
(0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel
|
146 |
+
(1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
|
147 |
+
)
|
148 |
+
```
|
149 |
+
|
150 |
+
## Usage
|
151 |
+
|
152 |
+
### Direct Usage (Sentence Transformers)
|
153 |
+
|
154 |
+
First install the Sentence Transformers library:
|
155 |
+
|
156 |
+
```bash
|
157 |
+
pip install -U sentence-transformers
|
158 |
+
```
|
159 |
+
|
160 |
+
Then you can load this model and run inference.
|
161 |
+
```python
|
162 |
+
from sentence_transformers import SentenceTransformer
|
163 |
+
|
164 |
+
# Download from the 🤗 Hub
|
165 |
+
model = SentenceTransformer("sentence_transformers_model_id")
|
166 |
+
# Run inference
|
167 |
+
sentences = [
|
168 |
+
'Ben vatansızım.',
|
169 |
+
'I am stateless.',
|
170 |
+
'Kendi tekniğini tercih ediyor.',
|
171 |
+
]
|
172 |
+
embeddings = model.encode(sentences)
|
173 |
+
print(embeddings.shape)
|
174 |
+
# [3, 384]
|
175 |
+
|
176 |
+
# Get the similarity scores for the embeddings
|
177 |
+
similarities = model.similarity(embeddings, embeddings)
|
178 |
+
print(similarities.shape)
|
179 |
+
# [3, 3]
|
180 |
+
```
|
181 |
+
|
182 |
+
<!--
|
183 |
+
### Direct Usage (Transformers)
|
184 |
+
|
185 |
+
<details><summary>Click to see the direct usage in Transformers</summary>
|
186 |
+
|
187 |
+
</details>
|
188 |
+
-->
|
189 |
+
|
190 |
+
<!--
|
191 |
+
### Downstream Usage (Sentence Transformers)
|
192 |
+
|
193 |
+
You can finetune this model on your own dataset.
|
194 |
+
|
195 |
+
<details><summary>Click to expand</summary>
|
196 |
+
|
197 |
+
</details>
|
198 |
+
-->
|
199 |
+
|
200 |
+
<!--
|
201 |
+
### Out-of-Scope Use
|
202 |
+
|
203 |
+
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
|
204 |
+
-->
|
205 |
+
|
206 |
+
## Evaluation
|
207 |
+
|
208 |
+
### Metrics
|
209 |
+
|
210 |
+
#### Semantic Similarity
|
211 |
+
* Dataset: `tr_ling`
|
212 |
+
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
|
213 |
+
|
214 |
+
| Metric | Value |
|
215 |
+
|:-------------------|:-----------|
|
216 |
+
| pearson_cosine | 0.0376 |
|
217 |
+
| spearman_cosine | 0.048 |
|
218 |
+
| pearson_manhattan | 0.0347 |
|
219 |
+
| spearman_manhattan | 0.0377 |
|
220 |
+
| pearson_euclidean | 0.037 |
|
221 |
+
| spearman_euclidean | 0.039 |
|
222 |
+
| pearson_dot | 0.0674 |
|
223 |
+
| spearman_dot | 0.0682 |
|
224 |
+
| pearson_max | 0.0674 |
|
225 |
+
| **spearman_max** | **0.0682** |
|
226 |
+
|
227 |
+
<!--
|
228 |
+
## Bias, Risks and Limitations
|
229 |
+
|
230 |
+
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
|
231 |
+
-->
|
232 |
+
|
233 |
+
<!--
|
234 |
+
### Recommendations
|
235 |
+
|
236 |
+
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
|
237 |
+
-->
|
238 |
+
|
239 |
+
## Training Details
|
240 |
+
|
241 |
+
### Training Dataset
|
242 |
+
|
243 |
+
#### MoritzLaurer/multilingual-nli-26lang-2mil7
|
244 |
+
|
245 |
+
* Dataset: [MoritzLaurer/multilingual-nli-26lang-2mil7](https://huggingface.co/datasets/MoritzLaurer/multilingual-nli-26lang-2mil7) at [510a233](https://huggingface.co/datasets/MoritzLaurer/multilingual-nli-26lang-2mil7/tree/510a233972a0d7ff0f767d82f46e046832c10538)
|
246 |
+
* Size: 25,000 training samples
|
247 |
+
* Columns: <code>premise_original</code>, <code>hypothesis_original</code>, <code>score</code>, <code>sentence1</code>, and <code>sentence2</code>
|
248 |
+
* Approximate statistics based on the first 1000 samples:
|
249 |
+
| | premise_original | hypothesis_original | score | sentence1 | sentence2 |
|
250 |
+
|:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
|
251 |
+
| type | string | string | int | string | string |
|
252 |
+
| details | <ul><li>min: 4 tokens</li><li>mean: 29.3 tokens</li><li>max: 107 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 15.62 tokens</li><li>max: 40 tokens</li></ul> | <ul><li>0: ~34.50%</li><li>1: ~33.30%</li><li>2: ~32.20%</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 28.28 tokens</li><li>max: 101 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 15.39 tokens</li><li>max: 38 tokens</li></ul> |
|
253 |
+
* Samples:
|
254 |
+
| premise_original | hypothesis_original | score | sentence1 | sentence2 |
|
255 |
+
|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------|
|
256 |
+
| <code>N, the total number of LC50 values used in calculating the CV(%) varied with organism and toxicant because some data were rejected due to water hardness, lack of concentration measurements, and/or because some of the LC50s were not calculable.</code> | <code>Most discarded data was rejected due to water hardness.</code> | <code>1</code> | <code>N, CV'nin hesaplanmasında kullanılan LC50 değerlerinin toplam sayısı (%) organizma ve toksik madde ile çeşitlidir, çünkü bazı veriler su sertliği, konsantrasyon ölçümlerinin eksikliği ve / veya LC50'lerin bazıları hesaplanamaz olduğu için reddedilmiştir.</code> | <code>Atılan verilerin çoğu su sertliği nedeniyle reddedildi.</code> |
|
257 |
+
| <code>As the home of the Venus de Milo and Mona Lisa, the Louvre drew almost unmanageable crowds until President Mitterrand ordered its re-organization in the 1980s.</code> | <code>The Louvre is home of the Venus de Milo and Mona Lisa.</code> | <code>0</code> | <code>Venus de Milo ve Mona Lisa'nın evi olarak Louvre, Başkan Mitterrand'ın 1980'lerde yeniden düzenlenmesini emredene kadar neredeyse yönetilemez kalabalıklar çekti.</code> | <code>Louvre, Venus de Milo ve Mona Lisa'nın evidir.</code> |
|
258 |
+
| <code>A year ago, the wife of the Oxford don noticed that the pattern on Kleenex quilted tissue uncannily resembled the Penrose Arrowed Rhombi tilings pattern, which Sir Roger had invented--and copyrighted--in 1974.</code> | <code>It has been recently found out a similarity between the pattern on the recent Kleenex quilted tissue and the one of the Penrose Arrowed Rhombi tilings.</code> | <code>0</code> | <code>Bir yıl önce Oxford'un karısı, Kleenex kapitone dokudaki desenin 1974'te Sir Roger'ın icat ettiği -ve telif hakkı olan - Penrose Arrowed Rhombi tilings desenine benzediğini fark etti.</code> | <code>Yakın zamanda, son Kleenex kapitone dokudaki desen ile Penrose Arrowed Rhombi döşemelerinden biri arasında bir benzerlik bulunmuştur.</code> |
|
259 |
+
* Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:
|
260 |
+
```json
|
261 |
+
{
|
262 |
+
"scale": 20.0,
|
263 |
+
"similarity_fct": "pairwise_cos_sim"
|
264 |
+
}
|
265 |
+
```
|
266 |
+
|
267 |
+
### Evaluation Dataset
|
268 |
+
|
269 |
+
#### MoritzLaurer/multilingual-nli-26lang-2mil7
|
270 |
+
|
271 |
+
* Dataset: [MoritzLaurer/multilingual-nli-26lang-2mil7](https://huggingface.co/datasets/MoritzLaurer/multilingual-nli-26lang-2mil7) at [510a233](https://huggingface.co/datasets/MoritzLaurer/multilingual-nli-26lang-2mil7/tree/510a233972a0d7ff0f767d82f46e046832c10538)
|
272 |
+
* Size: 5,000 evaluation samples
|
273 |
+
* Columns: <code>premise_original</code>, <code>hypothesis_original</code>, <code>score</code>, <code>sentence1</code>, and <code>sentence2</code>
|
274 |
+
* Approximate statistics based on the first 1000 samples:
|
275 |
+
| | premise_original | hypothesis_original | score | sentence1 | sentence2 |
|
276 |
+
|:--------|:---------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
|
277 |
+
| type | string | string | int | string | string |
|
278 |
+
| details | <ul><li>min: 5 tokens</li><li>mean: 30.3 tokens</li><li>max: 99 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 15.11 tokens</li><li>max: 56 tokens</li></ul> | <ul><li>0: ~34.50%</li><li>1: ~29.90%</li><li>2: ~35.60%</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 29.94 tokens</li><li>max: 106 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 15.29 tokens</li><li>max: 52 tokens</li></ul> |
|
279 |
+
* Samples:
|
280 |
+
| premise_original | hypothesis_original | score | sentence1 | sentence2 |
|
281 |
+
|:----------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------|:---------------|:------------------------------------------------------------------------------|:-----------------------------------------------------------------|
|
282 |
+
| <code>But the racism charge isn't quirky or wacky--it's demagogy.</code> | <code>The accusation of prejudice based on a pedestrian kind of hatred.</code> | <code>0</code> | <code>Ama ırkçılık suçlaması tuhaf ya da tuhaf değil, bu bir demagoji.</code> | <code>Yaya nefretine dayanan önyargı suçlaması.</code> |
|
283 |
+
| <code>Why would Gates allow the publication of such a book with his byline and photo on the dust jacket?</code> | <code>Gates' byline and photo are on the dust jacket</code> | <code>0</code> | <code>Gates neden böyle bir kitabın basılmasına izin versin ki?</code> | <code>Gates'in çizgisi ve fotoğrafı toz ceketin üzerinde.</code> |
|
284 |
+
| <code>I am a nonsmoker and allergic to cigarette smoke.</code> | <code>I do not smoke.</code> | <code>0</code> | <code>Sigara içmeyen biriyim ve sigara dumanına alerjim var.</code> | <code>Sigara içmiyorum.</code> |
|
285 |
+
* Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:
|
286 |
+
```json
|
287 |
+
{
|
288 |
+
"scale": 20.0,
|
289 |
+
"similarity_fct": "pairwise_cos_sim"
|
290 |
+
}
|
291 |
+
```
|
292 |
+
|
293 |
+
### Training Hyperparameters
|
294 |
+
#### Non-Default Hyperparameters
|
295 |
+
|
296 |
+
- `eval_strategy`: epoch
|
297 |
+
- `per_device_train_batch_size`: 32
|
298 |
+
- `per_device_eval_batch_size`: 64
|
299 |
+
- `learning_rate`: 2e-05
|
300 |
+
- `num_train_epochs`: 5
|
301 |
+
- `warmup_ratio`: 0.1
|
302 |
+
- `fp16`: True
|
303 |
+
- `load_best_model_at_end`: True
|
304 |
+
- `ddp_find_unused_parameters`: False
|
305 |
+
|
306 |
+
#### All Hyperparameters
|
307 |
+
<details><summary>Click to expand</summary>
|
308 |
+
|
309 |
+
- `overwrite_output_dir`: False
|
310 |
+
- `do_predict`: False
|
311 |
+
- `eval_strategy`: epoch
|
312 |
+
- `prediction_loss_only`: True
|
313 |
+
- `per_device_train_batch_size`: 32
|
314 |
+
- `per_device_eval_batch_size`: 64
|
315 |
+
- `per_gpu_train_batch_size`: None
|
316 |
+
- `per_gpu_eval_batch_size`: None
|
317 |
+
- `gradient_accumulation_steps`: 1
|
318 |
+
- `eval_accumulation_steps`: None
|
319 |
+
- `learning_rate`: 2e-05
|
320 |
+
- `weight_decay`: 0.0
|
321 |
+
- `adam_beta1`: 0.9
|
322 |
+
- `adam_beta2`: 0.999
|
323 |
+
- `adam_epsilon`: 1e-08
|
324 |
+
- `max_grad_norm`: 1.0
|
325 |
+
- `num_train_epochs`: 5
|
326 |
+
- `max_steps`: -1
|
327 |
+
- `lr_scheduler_type`: linear
|
328 |
+
- `lr_scheduler_kwargs`: {}
|
329 |
+
- `warmup_ratio`: 0.1
|
330 |
+
- `warmup_steps`: 0
|
331 |
+
- `log_level`: passive
|
332 |
+
- `log_level_replica`: warning
|
333 |
+
- `log_on_each_node`: True
|
334 |
+
- `logging_nan_inf_filter`: True
|
335 |
+
- `save_safetensors`: True
|
336 |
+
- `save_on_each_node`: False
|
337 |
+
- `save_only_model`: False
|
338 |
+
- `restore_callback_states_from_checkpoint`: False
|
339 |
+
- `no_cuda`: False
|
340 |
+
- `use_cpu`: False
|
341 |
+
- `use_mps_device`: False
|
342 |
+
- `seed`: 42
|
343 |
+
- `data_seed`: None
|
344 |
+
- `jit_mode_eval`: False
|
345 |
+
- `use_ipex`: False
|
346 |
+
- `bf16`: False
|
347 |
+
- `fp16`: True
|
348 |
+
- `fp16_opt_level`: O1
|
349 |
+
- `half_precision_backend`: auto
|
350 |
+
- `bf16_full_eval`: False
|
351 |
+
- `fp16_full_eval`: False
|
352 |
+
- `tf32`: None
|
353 |
+
- `local_rank`: 0
|
354 |
+
- `ddp_backend`: None
|
355 |
+
- `tpu_num_cores`: None
|
356 |
+
- `tpu_metrics_debug`: False
|
357 |
+
- `debug`: []
|
358 |
+
- `dataloader_drop_last`: False
|
359 |
+
- `dataloader_num_workers`: 0
|
360 |
+
- `dataloader_prefetch_factor`: None
|
361 |
+
- `past_index`: -1
|
362 |
+
- `disable_tqdm`: False
|
363 |
+
- `remove_unused_columns`: True
|
364 |
+
- `label_names`: None
|
365 |
+
- `load_best_model_at_end`: True
|
366 |
+
- `ignore_data_skip`: False
|
367 |
+
- `fsdp`: []
|
368 |
+
- `fsdp_min_num_params`: 0
|
369 |
+
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
|
370 |
+
- `fsdp_transformer_layer_cls_to_wrap`: None
|
371 |
+
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
|
372 |
+
- `deepspeed`: None
|
373 |
+
- `label_smoothing_factor`: 0.0
|
374 |
+
- `optim`: adamw_torch
|
375 |
+
- `optim_args`: None
|
376 |
+
- `adafactor`: False
|
377 |
+
- `group_by_length`: False
|
378 |
+
- `length_column_name`: length
|
379 |
+
- `ddp_find_unused_parameters`: False
|
380 |
+
- `ddp_bucket_cap_mb`: None
|
381 |
+
- `ddp_broadcast_buffers`: False
|
382 |
+
- `dataloader_pin_memory`: True
|
383 |
+
- `dataloader_persistent_workers`: False
|
384 |
+
- `skip_memory_metrics`: True
|
385 |
+
- `use_legacy_prediction_loop`: False
|
386 |
+
- `push_to_hub`: False
|
387 |
+
- `resume_from_checkpoint`: None
|
388 |
+
- `hub_model_id`: None
|
389 |
+
- `hub_strategy`: every_save
|
390 |
+
- `hub_private_repo`: False
|
391 |
+
- `hub_always_push`: False
|
392 |
+
- `gradient_checkpointing`: False
|
393 |
+
- `gradient_checkpointing_kwargs`: None
|
394 |
+
- `include_inputs_for_metrics`: False
|
395 |
+
- `eval_do_concat_batches`: True
|
396 |
+
- `fp16_backend`: auto
|
397 |
+
- `push_to_hub_model_id`: None
|
398 |
+
- `push_to_hub_organization`: None
|
399 |
+
- `mp_parameters`:
|
400 |
+
- `auto_find_batch_size`: False
|
401 |
+
- `full_determinism`: False
|
402 |
+
- `torchdynamo`: None
|
403 |
+
- `ray_scope`: last
|
404 |
+
- `ddp_timeout`: 1800
|
405 |
+
- `torch_compile`: False
|
406 |
+
- `torch_compile_backend`: None
|
407 |
+
- `torch_compile_mode`: None
|
408 |
+
- `dispatch_batches`: None
|
409 |
+
- `split_batches`: None
|
410 |
+
- `include_tokens_per_second`: False
|
411 |
+
- `include_num_input_tokens_seen`: False
|
412 |
+
- `neftune_noise_alpha`: None
|
413 |
+
- `optim_target_modules`: None
|
414 |
+
- `batch_eval_metrics`: False
|
415 |
+
- `batch_sampler`: batch_sampler
|
416 |
+
- `multi_dataset_batch_sampler`: proportional
|
417 |
+
|
418 |
+
</details>
|
419 |
+
|
420 |
+
### Training Logs
|
421 |
+
| Epoch | Step | Training Loss | loss | tr_ling_spearman_max |
|
422 |
+
|:------:|:----:|:-------------:|:------:|:--------------------:|
|
423 |
+
| 0.0320 | 25 | 17.17 | - | - |
|
424 |
+
| 0.0639 | 50 | 16.4932 | - | - |
|
425 |
+
| 0.0959 | 75 | 16.5976 | - | - |
|
426 |
+
| 0.1279 | 100 | 15.6991 | - | - |
|
427 |
+
| 0.1598 | 125 | 14.876 | - | - |
|
428 |
+
| 0.1918 | 150 | 14.4828 | - | - |
|
429 |
+
| 0.2238 | 175 | 12.7061 | - | - |
|
430 |
+
| 0.2558 | 200 | 10.8687 | - | - |
|
431 |
+
| 0.2877 | 225 | 8.3797 | - | - |
|
432 |
+
| 0.3197 | 250 | 6.2029 | - | - |
|
433 |
+
| 0.3517 | 275 | 5.8228 | - | - |
|
434 |
+
| 0.3836 | 300 | 5.811 | - | - |
|
435 |
+
| 0.4156 | 325 | 5.8079 | - | - |
|
436 |
+
| 0.4476 | 350 | 5.8077 | - | - |
|
437 |
+
| 0.4795 | 375 | 5.8035 | - | - |
|
438 |
+
| 0.5115 | 400 | 5.8072 | - | - |
|
439 |
+
| 0.5435 | 425 | 5.8033 | - | - |
|
440 |
+
| 0.5754 | 450 | 5.8086 | - | - |
|
441 |
+
| 0.6074 | 475 | 5.81 | - | - |
|
442 |
+
| 0.6394 | 500 | 5.7949 | - | - |
|
443 |
+
| 0.6714 | 525 | 5.8079 | - | - |
|
444 |
+
| 0.7033 | 550 | 5.8057 | - | - |
|
445 |
+
| 0.7353 | 575 | 5.8097 | - | - |
|
446 |
+
| 0.7673 | 600 | 5.7986 | - | - |
|
447 |
+
| 0.7992 | 625 | 5.8051 | - | - |
|
448 |
+
| 0.8312 | 650 | 5.8041 | - | - |
|
449 |
+
| 0.8632 | 675 | 5.7907 | - | - |
|
450 |
+
| 0.8951 | 700 | 5.7991 | - | - |
|
451 |
+
| 0.9271 | 725 | 5.8035 | - | - |
|
452 |
+
| 0.9591 | 750 | 5.7945 | - | - |
|
453 |
+
| 0.9910 | 775 | 5.8077 | - | - |
|
454 |
+
| 1.0 | 782 | - | 5.8024 | 0.0330 |
|
455 |
+
| 1.0230 | 800 | 5.6703 | - | - |
|
456 |
+
| 1.0550 | 825 | 5.8052 | - | - |
|
457 |
+
| 1.0870 | 850 | 5.7936 | - | - |
|
458 |
+
| 1.1189 | 875 | 5.7924 | - | - |
|
459 |
+
| 1.1509 | 900 | 5.7806 | - | - |
|
460 |
+
| 1.1829 | 925 | 5.7835 | - | - |
|
461 |
+
| 1.2148 | 950 | 5.7619 | - | - |
|
462 |
+
| 1.2468 | 975 | 5.8038 | - | - |
|
463 |
+
| 1.2788 | 1000 | 5.779 | - | - |
|
464 |
+
| 1.3107 | 1025 | 5.7904 | - | - |
|
465 |
+
| 1.3427 | 1050 | 5.7696 | - | - |
|
466 |
+
| 1.3747 | 1075 | 5.7919 | - | - |
|
467 |
+
| 1.4066 | 1100 | 5.7785 | - | - |
|
468 |
+
| 1.4386 | 1125 | 5.7862 | - | - |
|
469 |
+
| 1.4706 | 1150 | 5.7703 | - | - |
|
470 |
+
| 1.5026 | 1175 | 5.773 | - | - |
|
471 |
+
| 1.5345 | 1200 | 5.7627 | - | - |
|
472 |
+
| 1.5665 | 1225 | 5.7596 | - | - |
|
473 |
+
| 1.5985 | 1250 | 5.7882 | - | - |
|
474 |
+
| 1.6304 | 1275 | 5.7828 | - | - |
|
475 |
+
| 1.6624 | 1300 | 5.771 | - | - |
|
476 |
+
| 1.6944 | 1325 | 5.788 | - | - |
|
477 |
+
| 1.7263 | 1350 | 5.7719 | - | - |
|
478 |
+
| 1.7583 | 1375 | 5.7846 | - | - |
|
479 |
+
| 1.7903 | 1400 | 5.7838 | - | - |
|
480 |
+
| 1.8223 | 1425 | 5.7912 | - | - |
|
481 |
+
| 1.8542 | 1450 | 5.7686 | - | - |
|
482 |
+
| 1.8862 | 1475 | 5.7938 | - | - |
|
483 |
+
| 1.9182 | 1500 | 5.7847 | - | - |
|
484 |
+
| 1.9501 | 1525 | 5.7952 | - | - |
|
485 |
+
| 1.9821 | 1550 | 5.7528 | - | - |
|
486 |
+
| 2.0 | 1564 | - | 5.7933 | 0.0682 |
|
487 |
+
|
488 |
+
|
489 |
+
### Framework Versions
|
490 |
+
- Python: 3.10.12
|
491 |
+
- Sentence Transformers: 3.0.0
|
492 |
+
- Transformers: 4.41.0
|
493 |
+
- PyTorch: 2.3.0+cu121
|
494 |
+
- Accelerate: 0.30.1
|
495 |
+
- Datasets: 2.19.1
|
496 |
+
- Tokenizers: 0.19.1
|
497 |
+
|
498 |
+
## Citation
|
499 |
+
|
500 |
+
### BibTeX
|
501 |
+
|
502 |
+
#### Sentence Transformers
|
503 |
+
```bibtex
|
504 |
+
@inproceedings{reimers-2019-sentence-bert,
|
505 |
+
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
|
506 |
+
author = "Reimers, Nils and Gurevych, Iryna",
|
507 |
+
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
|
508 |
+
month = "11",
|
509 |
+
year = "2019",
|
510 |
+
publisher = "Association for Computational Linguistics",
|
511 |
+
url = "https://arxiv.org/abs/1908.10084",
|
512 |
+
}
|
513 |
+
```
|
514 |
+
|
515 |
+
#### CoSENTLoss
|
516 |
+
```bibtex
|
517 |
+
@online{kexuefm-8847,
|
518 |
+
title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
|
519 |
+
author={Su Jianlin},
|
520 |
+
year={2022},
|
521 |
+
month={Jan},
|
522 |
+
url={https://kexue.fm/archives/8847},
|
523 |
+
}
|
524 |
+
```
|
525 |
+
|
526 |
+
<!--
|
527 |
+
## Glossary
|
528 |
+
|
529 |
+
*Clearly define terms in order to be accessible across audiences.*
|
530 |
+
-->
|
531 |
+
|
532 |
+
<!--
|
533 |
+
## Model Card Authors
|
534 |
+
|
535 |
+
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
|
536 |
+
-->
|
537 |
+
|
538 |
+
<!--
|
539 |
+
## Model Card Contact
|
540 |
+
|
541 |
+
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
|
542 |
+
-->
|
checkpoint-1564/config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2",
|
3 |
+
"architectures": [
|
4 |
+
"BertModel"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"classifier_dropout": null,
|
8 |
+
"gradient_checkpointing": false,
|
9 |
+
"hidden_act": "gelu",
|
10 |
+
"hidden_dropout_prob": 0.1,
|
11 |
+
"hidden_size": 384,
|
12 |
+
"initializer_range": 0.02,
|
13 |
+
"intermediate_size": 1536,
|
14 |
+
"layer_norm_eps": 1e-12,
|
15 |
+
"max_position_embeddings": 512,
|
16 |
+
"model_type": "bert",
|
17 |
+
"num_attention_heads": 12,
|
18 |
+
"num_hidden_layers": 12,
|
19 |
+
"pad_token_id": 0,
|
20 |
+
"position_embedding_type": "absolute",
|
21 |
+
"torch_dtype": "float32",
|
22 |
+
"transformers_version": "4.41.0",
|
23 |
+
"type_vocab_size": 2,
|
24 |
+
"use_cache": true,
|
25 |
+
"vocab_size": 250037
|
26 |
+
}
|
checkpoint-1564/config_sentence_transformers.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"__version__": {
|
3 |
+
"sentence_transformers": "2.0.0",
|
4 |
+
"transformers": "4.7.0",
|
5 |
+
"pytorch": "1.9.0+cu102"
|
6 |
+
},
|
7 |
+
"prompts": {},
|
8 |
+
"default_prompt_name": null,
|
9 |
+
"similarity_fn_name": null
|
10 |
+
}
|
checkpoint-1564/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7a3b103c4a050ec7c20c41f4f0bdcfee82642ecfe0cd58c6268c2bfddce5abbd
|
3 |
+
size 470637416
|
checkpoint-1564/modules.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
[
|
2 |
+
{
|
3 |
+
"idx": 0,
|
4 |
+
"name": "0",
|
5 |
+
"path": "",
|
6 |
+
"type": "sentence_transformers.models.Transformer"
|
7 |
+
},
|
8 |
+
{
|
9 |
+
"idx": 1,
|
10 |
+
"name": "1",
|
11 |
+
"path": "1_Pooling",
|
12 |
+
"type": "sentence_transformers.models.Pooling"
|
13 |
+
}
|
14 |
+
]
|
checkpoint-1564/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a4fd1f2073e2af85a9380b42823863bdcd018438654cbfa5b3aa98b918223855
|
3 |
+
size 940212218
|
checkpoint-1564/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:87c6bc4d1d198c376a51fede7dca1cc0874e21a3307b8ae3875e98db55be87e6
|
3 |
+
size 14244
|
checkpoint-1564/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fcc21c2650bb4344f4670f8f3de62ba81c0b68e93028800e195846665d9176d9
|
3 |
+
size 1064
|
checkpoint-1564/sentence_bert_config.json
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"max_seq_length": 128,
|
3 |
+
"do_lower_case": false
|
4 |
+
}
|
checkpoint-1564/special_tokens_map.json
ADDED
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<s>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"cls_token": {
|
10 |
+
"content": "<s>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"eos_token": {
|
17 |
+
"content": "</s>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
},
|
23 |
+
"mask_token": {
|
24 |
+
"content": "<mask>",
|
25 |
+
"lstrip": true,
|
26 |
+
"normalized": false,
|
27 |
+
"rstrip": false,
|
28 |
+
"single_word": false
|
29 |
+
},
|
30 |
+
"pad_token": {
|
31 |
+
"content": "<pad>",
|
32 |
+
"lstrip": false,
|
33 |
+
"normalized": false,
|
34 |
+
"rstrip": false,
|
35 |
+
"single_word": false
|
36 |
+
},
|
37 |
+
"sep_token": {
|
38 |
+
"content": "</s>",
|
39 |
+
"lstrip": false,
|
40 |
+
"normalized": false,
|
41 |
+
"rstrip": false,
|
42 |
+
"single_word": false
|
43 |
+
},
|
44 |
+
"unk_token": {
|
45 |
+
"content": "<unk>",
|
46 |
+
"lstrip": false,
|
47 |
+
"normalized": false,
|
48 |
+
"rstrip": false,
|
49 |
+
"single_word": false
|
50 |
+
}
|
51 |
+
}
|
checkpoint-1564/tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cad551d5600a84242d0973327029452a1e3672ba6313c2a3c3d69c4310e12719
|
3 |
+
size 17082987
|
checkpoint-1564/tokenizer_config.json
ADDED
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "<s>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "<pad>",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "</s>",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"3": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"250001": {
|
36 |
+
"content": "<mask>",
|
37 |
+
"lstrip": true,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
}
|
43 |
+
},
|
44 |
+
"bos_token": "<s>",
|
45 |
+
"clean_up_tokenization_spaces": true,
|
46 |
+
"cls_token": "<s>",
|
47 |
+
"do_lower_case": true,
|
48 |
+
"eos_token": "</s>",
|
49 |
+
"mask_token": "<mask>",
|
50 |
+
"max_length": 128,
|
51 |
+
"model_max_length": 128,
|
52 |
+
"pad_to_multiple_of": null,
|
53 |
+
"pad_token": "<pad>",
|
54 |
+
"pad_token_type_id": 0,
|
55 |
+
"padding_side": "right",
|
56 |
+
"sep_token": "</s>",
|
57 |
+
"stride": 0,
|
58 |
+
"strip_accents": null,
|
59 |
+
"tokenize_chinese_chars": true,
|
60 |
+
"tokenizer_class": "BertTokenizer",
|
61 |
+
"truncation_side": "right",
|
62 |
+
"truncation_strategy": "longest_first",
|
63 |
+
"unk_token": "<unk>"
|
64 |
+
}
|
checkpoint-1564/trainer_state.json
ADDED
@@ -0,0 +1,512 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 5.793323516845703,
|
3 |
+
"best_model_checkpoint": "turkish-embedding-model/checkpoint-1564",
|
4 |
+
"epoch": 2.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1564,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0319693094629156,
|
13 |
+
"grad_norm": 70.72708129882812,
|
14 |
+
"learning_rate": 1.1253196930946293e-06,
|
15 |
+
"loss": 17.17,
|
16 |
+
"step": 25
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.0639386189258312,
|
20 |
+
"grad_norm": 81.68770599365234,
|
21 |
+
"learning_rate": 2.4040920716112534e-06,
|
22 |
+
"loss": 16.4932,
|
23 |
+
"step": 50
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.0959079283887468,
|
27 |
+
"grad_norm": 109.91338348388672,
|
28 |
+
"learning_rate": 3.6828644501278778e-06,
|
29 |
+
"loss": 16.5976,
|
30 |
+
"step": 75
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.1278772378516624,
|
34 |
+
"grad_norm": 73.892578125,
|
35 |
+
"learning_rate": 4.961636828644502e-06,
|
36 |
+
"loss": 15.6991,
|
37 |
+
"step": 100
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.159846547314578,
|
41 |
+
"grad_norm": 79.35150909423828,
|
42 |
+
"learning_rate": 6.240409207161126e-06,
|
43 |
+
"loss": 14.876,
|
44 |
+
"step": 125
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.1918158567774936,
|
48 |
+
"grad_norm": 83.0904541015625,
|
49 |
+
"learning_rate": 7.5191815856777495e-06,
|
50 |
+
"loss": 14.4828,
|
51 |
+
"step": 150
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.2237851662404092,
|
55 |
+
"grad_norm": 76.82855987548828,
|
56 |
+
"learning_rate": 8.797953964194374e-06,
|
57 |
+
"loss": 12.7061,
|
58 |
+
"step": 175
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.2557544757033248,
|
62 |
+
"grad_norm": 51.30181121826172,
|
63 |
+
"learning_rate": 1.0076726342710998e-05,
|
64 |
+
"loss": 10.8687,
|
65 |
+
"step": 200
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.2877237851662404,
|
69 |
+
"grad_norm": 18.70808219909668,
|
70 |
+
"learning_rate": 1.1355498721227622e-05,
|
71 |
+
"loss": 8.3797,
|
72 |
+
"step": 225
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.319693094629156,
|
76 |
+
"grad_norm": 1.3039417266845703,
|
77 |
+
"learning_rate": 1.2634271099744246e-05,
|
78 |
+
"loss": 6.2029,
|
79 |
+
"step": 250
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.3516624040920716,
|
83 |
+
"grad_norm": 0.2324853092432022,
|
84 |
+
"learning_rate": 1.391304347826087e-05,
|
85 |
+
"loss": 5.8228,
|
86 |
+
"step": 275
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.3836317135549872,
|
90 |
+
"grad_norm": 0.1757364720106125,
|
91 |
+
"learning_rate": 1.5191815856777494e-05,
|
92 |
+
"loss": 5.811,
|
93 |
+
"step": 300
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.4156010230179028,
|
97 |
+
"grad_norm": 0.1788654774427414,
|
98 |
+
"learning_rate": 1.647058823529412e-05,
|
99 |
+
"loss": 5.8079,
|
100 |
+
"step": 325
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.4475703324808184,
|
104 |
+
"grad_norm": 0.12862567603588104,
|
105 |
+
"learning_rate": 1.7749360613810744e-05,
|
106 |
+
"loss": 5.8077,
|
107 |
+
"step": 350
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.479539641943734,
|
111 |
+
"grad_norm": 0.14497514069080353,
|
112 |
+
"learning_rate": 1.9028132992327367e-05,
|
113 |
+
"loss": 5.8035,
|
114 |
+
"step": 375
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.5115089514066496,
|
118 |
+
"grad_norm": 0.1350390762090683,
|
119 |
+
"learning_rate": 1.996589940323956e-05,
|
120 |
+
"loss": 5.8072,
|
121 |
+
"step": 400
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.5434782608695652,
|
125 |
+
"grad_norm": 0.1435602754354477,
|
126 |
+
"learning_rate": 1.9823813583404378e-05,
|
127 |
+
"loss": 5.8033,
|
128 |
+
"step": 425
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.5754475703324808,
|
132 |
+
"grad_norm": 0.11389254033565521,
|
133 |
+
"learning_rate": 1.96817277635692e-05,
|
134 |
+
"loss": 5.8086,
|
135 |
+
"step": 450
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.6074168797953964,
|
139 |
+
"grad_norm": 0.15821650624275208,
|
140 |
+
"learning_rate": 1.9539641943734017e-05,
|
141 |
+
"loss": 5.81,
|
142 |
+
"step": 475
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.639386189258312,
|
146 |
+
"grad_norm": 0.1179889366030693,
|
147 |
+
"learning_rate": 1.9397556123898838e-05,
|
148 |
+
"loss": 5.7949,
|
149 |
+
"step": 500
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.6713554987212276,
|
153 |
+
"grad_norm": 0.10912967473268509,
|
154 |
+
"learning_rate": 1.9255470304063656e-05,
|
155 |
+
"loss": 5.8079,
|
156 |
+
"step": 525
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.7033248081841432,
|
160 |
+
"grad_norm": 0.11702870577573776,
|
161 |
+
"learning_rate": 1.9113384484228477e-05,
|
162 |
+
"loss": 5.8057,
|
163 |
+
"step": 550
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.7352941176470589,
|
167 |
+
"grad_norm": 0.13132448494434357,
|
168 |
+
"learning_rate": 1.8971298664393295e-05,
|
169 |
+
"loss": 5.8097,
|
170 |
+
"step": 575
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.7672634271099744,
|
174 |
+
"grad_norm": 0.15833145380020142,
|
175 |
+
"learning_rate": 1.8829212844558116e-05,
|
176 |
+
"loss": 5.7986,
|
177 |
+
"step": 600
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.7992327365728901,
|
181 |
+
"grad_norm": 0.11651863902807236,
|
182 |
+
"learning_rate": 1.8687127024722937e-05,
|
183 |
+
"loss": 5.8051,
|
184 |
+
"step": 625
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.8312020460358056,
|
188 |
+
"grad_norm": 0.5393890142440796,
|
189 |
+
"learning_rate": 1.854504120488775e-05,
|
190 |
+
"loss": 5.8041,
|
191 |
+
"step": 650
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.8631713554987213,
|
195 |
+
"grad_norm": 0.6457561254501343,
|
196 |
+
"learning_rate": 1.8402955385052572e-05,
|
197 |
+
"loss": 5.7907,
|
198 |
+
"step": 675
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.8951406649616368,
|
202 |
+
"grad_norm": 0.5643135905265808,
|
203 |
+
"learning_rate": 1.8260869565217393e-05,
|
204 |
+
"loss": 5.7991,
|
205 |
+
"step": 700
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.9271099744245525,
|
209 |
+
"grad_norm": 3.214787721633911,
|
210 |
+
"learning_rate": 1.811878374538221e-05,
|
211 |
+
"loss": 5.8035,
|
212 |
+
"step": 725
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.959079283887468,
|
216 |
+
"grad_norm": 2.781162977218628,
|
217 |
+
"learning_rate": 1.7976697925547032e-05,
|
218 |
+
"loss": 5.7945,
|
219 |
+
"step": 750
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.9910485933503836,
|
223 |
+
"grad_norm": 0.38559335470199585,
|
224 |
+
"learning_rate": 1.783461210571185e-05,
|
225 |
+
"loss": 5.8077,
|
226 |
+
"step": 775
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 1.0,
|
230 |
+
"eval_loss": 5.8023600578308105,
|
231 |
+
"eval_runtime": 18.0632,
|
232 |
+
"eval_samples_per_second": 276.805,
|
233 |
+
"eval_steps_per_second": 4.374,
|
234 |
+
"eval_tr_ling_pearson_cosine": 0.017751548525136808,
|
235 |
+
"eval_tr_ling_pearson_dot": 0.025703597820631346,
|
236 |
+
"eval_tr_ling_pearson_euclidean": 0.02195284877201089,
|
237 |
+
"eval_tr_ling_pearson_manhattan": 0.02083376479528459,
|
238 |
+
"eval_tr_ling_pearson_max": 0.025703597820631346,
|
239 |
+
"eval_tr_ling_spearman_cosine": 0.027108099994157316,
|
240 |
+
"eval_tr_ling_spearman_dot": 0.03304394653738539,
|
241 |
+
"eval_tr_ling_spearman_euclidean": 0.025485959636772793,
|
242 |
+
"eval_tr_ling_spearman_manhattan": 0.024466610177699702,
|
243 |
+
"eval_tr_ling_spearman_max": 0.03304394653738539,
|
244 |
+
"step": 782
|
245 |
+
},
|
246 |
+
{
|
247 |
+
"epoch": 1.0230179028132993,
|
248 |
+
"grad_norm": 0.3645063638687134,
|
249 |
+
"learning_rate": 1.769252628587667e-05,
|
250 |
+
"loss": 5.6703,
|
251 |
+
"step": 800
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 1.054987212276215,
|
255 |
+
"grad_norm": 0.9638137817382812,
|
256 |
+
"learning_rate": 1.7550440466041488e-05,
|
257 |
+
"loss": 5.8052,
|
258 |
+
"step": 825
|
259 |
+
},
|
260 |
+
{
|
261 |
+
"epoch": 1.0869565217391304,
|
262 |
+
"grad_norm": 2.114203691482544,
|
263 |
+
"learning_rate": 1.740835464620631e-05,
|
264 |
+
"loss": 5.7936,
|
265 |
+
"step": 850
|
266 |
+
},
|
267 |
+
{
|
268 |
+
"epoch": 1.118925831202046,
|
269 |
+
"grad_norm": 1.8992066383361816,
|
270 |
+
"learning_rate": 1.7266268826371127e-05,
|
271 |
+
"loss": 5.7924,
|
272 |
+
"step": 875
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 1.1508951406649617,
|
276 |
+
"grad_norm": 2.8299577236175537,
|
277 |
+
"learning_rate": 1.7124183006535948e-05,
|
278 |
+
"loss": 5.7806,
|
279 |
+
"step": 900
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 1.1828644501278773,
|
283 |
+
"grad_norm": 1.956953525543213,
|
284 |
+
"learning_rate": 1.698209718670077e-05,
|
285 |
+
"loss": 5.7835,
|
286 |
+
"step": 925
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 1.2148337595907928,
|
290 |
+
"grad_norm": 2.658413887023926,
|
291 |
+
"learning_rate": 1.6840011366865587e-05,
|
292 |
+
"loss": 5.7619,
|
293 |
+
"step": 950
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 1.2468030690537084,
|
297 |
+
"grad_norm": 1.2760388851165771,
|
298 |
+
"learning_rate": 1.6697925547030408e-05,
|
299 |
+
"loss": 5.8038,
|
300 |
+
"step": 975
|
301 |
+
},
|
302 |
+
{
|
303 |
+
"epoch": 1.278772378516624,
|
304 |
+
"grad_norm": 1.7434897422790527,
|
305 |
+
"learning_rate": 1.6555839727195226e-05,
|
306 |
+
"loss": 5.779,
|
307 |
+
"step": 1000
|
308 |
+
},
|
309 |
+
{
|
310 |
+
"epoch": 1.3107416879795397,
|
311 |
+
"grad_norm": 1.3532071113586426,
|
312 |
+
"learning_rate": 1.6413753907360047e-05,
|
313 |
+
"loss": 5.7904,
|
314 |
+
"step": 1025
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 1.3427109974424551,
|
318 |
+
"grad_norm": 3.7385997772216797,
|
319 |
+
"learning_rate": 1.6271668087524864e-05,
|
320 |
+
"loss": 5.7696,
|
321 |
+
"step": 1050
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 1.3746803069053708,
|
325 |
+
"grad_norm": 0.9061102867126465,
|
326 |
+
"learning_rate": 1.6129582267689685e-05,
|
327 |
+
"loss": 5.7919,
|
328 |
+
"step": 1075
|
329 |
+
},
|
330 |
+
{
|
331 |
+
"epoch": 1.4066496163682864,
|
332 |
+
"grad_norm": 2.7104809284210205,
|
333 |
+
"learning_rate": 1.5987496447854503e-05,
|
334 |
+
"loss": 5.7785,
|
335 |
+
"step": 1100
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 1.438618925831202,
|
339 |
+
"grad_norm": 1.7147830724716187,
|
340 |
+
"learning_rate": 1.5845410628019324e-05,
|
341 |
+
"loss": 5.7862,
|
342 |
+
"step": 1125
|
343 |
+
},
|
344 |
+
{
|
345 |
+
"epoch": 1.4705882352941178,
|
346 |
+
"grad_norm": 2.525214672088623,
|
347 |
+
"learning_rate": 1.5703324808184145e-05,
|
348 |
+
"loss": 5.7703,
|
349 |
+
"step": 1150
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 1.5025575447570332,
|
353 |
+
"grad_norm": 1.7794997692108154,
|
354 |
+
"learning_rate": 1.5561238988348963e-05,
|
355 |
+
"loss": 5.773,
|
356 |
+
"step": 1175
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 1.5345268542199488,
|
360 |
+
"grad_norm": 4.901644229888916,
|
361 |
+
"learning_rate": 1.5419153168513784e-05,
|
362 |
+
"loss": 5.7627,
|
363 |
+
"step": 1200
|
364 |
+
},
|
365 |
+
{
|
366 |
+
"epoch": 1.5664961636828645,
|
367 |
+
"grad_norm": 3.360812187194824,
|
368 |
+
"learning_rate": 1.52770673486786e-05,
|
369 |
+
"loss": 5.7596,
|
370 |
+
"step": 1225
|
371 |
+
},
|
372 |
+
{
|
373 |
+
"epoch": 1.59846547314578,
|
374 |
+
"grad_norm": 1.2768888473510742,
|
375 |
+
"learning_rate": 1.5134981528843423e-05,
|
376 |
+
"loss": 5.7882,
|
377 |
+
"step": 1250
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 1.6304347826086958,
|
381 |
+
"grad_norm": 2.206226348876953,
|
382 |
+
"learning_rate": 1.4992895709008242e-05,
|
383 |
+
"loss": 5.7828,
|
384 |
+
"step": 1275
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 1.6624040920716112,
|
388 |
+
"grad_norm": 1.4602406024932861,
|
389 |
+
"learning_rate": 1.4850809889173061e-05,
|
390 |
+
"loss": 5.771,
|
391 |
+
"step": 1300
|
392 |
+
},
|
393 |
+
{
|
394 |
+
"epoch": 1.6943734015345269,
|
395 |
+
"grad_norm": 1.1597537994384766,
|
396 |
+
"learning_rate": 1.4708724069337881e-05,
|
397 |
+
"loss": 5.788,
|
398 |
+
"step": 1325
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 1.7263427109974425,
|
402 |
+
"grad_norm": 3.7494003772735596,
|
403 |
+
"learning_rate": 1.45666382495027e-05,
|
404 |
+
"loss": 5.7719,
|
405 |
+
"step": 1350
|
406 |
+
},
|
407 |
+
{
|
408 |
+
"epoch": 1.758312020460358,
|
409 |
+
"grad_norm": 1.6271498203277588,
|
410 |
+
"learning_rate": 1.442455242966752e-05,
|
411 |
+
"loss": 5.7846,
|
412 |
+
"step": 1375
|
413 |
+
},
|
414 |
+
{
|
415 |
+
"epoch": 1.7902813299232738,
|
416 |
+
"grad_norm": 2.0469117164611816,
|
417 |
+
"learning_rate": 1.4282466609832339e-05,
|
418 |
+
"loss": 5.7838,
|
419 |
+
"step": 1400
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 1.8222506393861893,
|
423 |
+
"grad_norm": 2.533921003341675,
|
424 |
+
"learning_rate": 1.4140380789997158e-05,
|
425 |
+
"loss": 5.7912,
|
426 |
+
"step": 1425
|
427 |
+
},
|
428 |
+
{
|
429 |
+
"epoch": 1.854219948849105,
|
430 |
+
"grad_norm": 3.291757583618164,
|
431 |
+
"learning_rate": 1.3998294970161978e-05,
|
432 |
+
"loss": 5.7686,
|
433 |
+
"step": 1450
|
434 |
+
},
|
435 |
+
{
|
436 |
+
"epoch": 1.8861892583120206,
|
437 |
+
"grad_norm": 3.0181350708007812,
|
438 |
+
"learning_rate": 1.3856209150326799e-05,
|
439 |
+
"loss": 5.7938,
|
440 |
+
"step": 1475
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 1.918158567774936,
|
444 |
+
"grad_norm": 2.553502321243286,
|
445 |
+
"learning_rate": 1.3714123330491618e-05,
|
446 |
+
"loss": 5.7847,
|
447 |
+
"step": 1500
|
448 |
+
},
|
449 |
+
{
|
450 |
+
"epoch": 1.9501278772378516,
|
451 |
+
"grad_norm": 1.8034719228744507,
|
452 |
+
"learning_rate": 1.3572037510656438e-05,
|
453 |
+
"loss": 5.7952,
|
454 |
+
"step": 1525
|
455 |
+
},
|
456 |
+
{
|
457 |
+
"epoch": 1.9820971867007673,
|
458 |
+
"grad_norm": 3.7138864994049072,
|
459 |
+
"learning_rate": 1.3429951690821257e-05,
|
460 |
+
"loss": 5.7528,
|
461 |
+
"step": 1550
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 2.0,
|
465 |
+
"eval_loss": 5.793323516845703,
|
466 |
+
"eval_runtime": 18.2796,
|
467 |
+
"eval_samples_per_second": 273.528,
|
468 |
+
"eval_steps_per_second": 4.322,
|
469 |
+
"eval_tr_ling_pearson_cosine": 0.037604255015168134,
|
470 |
+
"eval_tr_ling_pearson_dot": 0.0673696846368413,
|
471 |
+
"eval_tr_ling_pearson_euclidean": 0.03698411306484619,
|
472 |
+
"eval_tr_ling_pearson_manhattan": 0.034740275152181296,
|
473 |
+
"eval_tr_ling_pearson_max": 0.0673696846368413,
|
474 |
+
"eval_tr_ling_spearman_cosine": 0.04804112988506346,
|
475 |
+
"eval_tr_ling_spearman_dot": 0.06818119362900125,
|
476 |
+
"eval_tr_ling_spearman_euclidean": 0.03903062430281842,
|
477 |
+
"eval_tr_ling_spearman_manhattan": 0.03769766156967754,
|
478 |
+
"eval_tr_ling_spearman_max": 0.06818119362900125,
|
479 |
+
"step": 1564
|
480 |
+
}
|
481 |
+
],
|
482 |
+
"logging_steps": 25,
|
483 |
+
"max_steps": 3910,
|
484 |
+
"num_input_tokens_seen": 0,
|
485 |
+
"num_train_epochs": 5,
|
486 |
+
"save_steps": 500,
|
487 |
+
"stateful_callbacks": {
|
488 |
+
"EarlyStoppingCallback": {
|
489 |
+
"args": {
|
490 |
+
"early_stopping_patience": 5,
|
491 |
+
"early_stopping_threshold": 0.01
|
492 |
+
},
|
493 |
+
"attributes": {
|
494 |
+
"early_stopping_patience_counter": 0
|
495 |
+
}
|
496 |
+
},
|
497 |
+
"TrainerControl": {
|
498 |
+
"args": {
|
499 |
+
"should_epoch_stop": false,
|
500 |
+
"should_evaluate": false,
|
501 |
+
"should_log": false,
|
502 |
+
"should_save": true,
|
503 |
+
"should_training_stop": false
|
504 |
+
},
|
505 |
+
"attributes": {}
|
506 |
+
}
|
507 |
+
},
|
508 |
+
"total_flos": 0.0,
|
509 |
+
"train_batch_size": 32,
|
510 |
+
"trial_name": null,
|
511 |
+
"trial_params": null
|
512 |
+
}
|
checkpoint-1564/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d44b0bec8869c08bdad0d597184d7a293a0e13eb770d6f4384456cbbe4fe5aa4
|
3 |
+
size 5368
|
checkpoint-1564/unigram.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:da145b5e7700ae40f16691ec32a0b1fdc1ee3298db22a31ea55f57a966c4a65d
|
3 |
+
size 14763260
|
checkpoint-3128/1_Pooling/config.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"word_embedding_dimension": 384,
|
3 |
+
"pooling_mode_cls_token": false,
|
4 |
+
"pooling_mode_mean_tokens": true,
|
5 |
+
"pooling_mode_max_tokens": false,
|
6 |
+
"pooling_mode_mean_sqrt_len_tokens": false,
|
7 |
+
"pooling_mode_weightedmean_tokens": false,
|
8 |
+
"pooling_mode_lasttoken": false,
|
9 |
+
"include_prompt": true
|
10 |
+
}
|
checkpoint-3128/README.md
ADDED
@@ -0,0 +1,610 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- multilingual
|
4 |
+
- zh
|
5 |
+
- ja
|
6 |
+
- ar
|
7 |
+
- ko
|
8 |
+
- de
|
9 |
+
- fr
|
10 |
+
- es
|
11 |
+
- pt
|
12 |
+
- hi
|
13 |
+
- id
|
14 |
+
- it
|
15 |
+
- tr
|
16 |
+
- ru
|
17 |
+
- bn
|
18 |
+
- ur
|
19 |
+
- mr
|
20 |
+
- ta
|
21 |
+
- vi
|
22 |
+
- fa
|
23 |
+
- pl
|
24 |
+
- uk
|
25 |
+
- nl
|
26 |
+
- sv
|
27 |
+
- he
|
28 |
+
- sw
|
29 |
+
- ps
|
30 |
+
library_name: sentence-transformers
|
31 |
+
tags:
|
32 |
+
- sentence-transformers
|
33 |
+
- sentence-similarity
|
34 |
+
- feature-extraction
|
35 |
+
- dataset_size:10K<n<100K
|
36 |
+
- loss:CoSENTLoss
|
37 |
+
base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
|
38 |
+
metrics:
|
39 |
+
- pearson_cosine
|
40 |
+
- spearman_cosine
|
41 |
+
- pearson_manhattan
|
42 |
+
- spearman_manhattan
|
43 |
+
- pearson_euclidean
|
44 |
+
- spearman_euclidean
|
45 |
+
- pearson_dot
|
46 |
+
- spearman_dot
|
47 |
+
- pearson_max
|
48 |
+
- spearman_max
|
49 |
+
widget:
|
50 |
+
- source_sentence: Bottomless Mug
|
51 |
+
sentences:
|
52 |
+
- You are always safe.
|
53 |
+
- That trend isn't very known yet
|
54 |
+
- Eleanor Clift göreve koşuyor.
|
55 |
+
- source_sentence: Tripp has a job.
|
56 |
+
sentences:
|
57 |
+
- They are having money problems.
|
58 |
+
- Malignite aniden ortaya çıkar.
|
59 |
+
- Mezarlar derin ormanlarda saklandı.
|
60 |
+
- source_sentence: There are rules
|
61 |
+
sentences:
|
62 |
+
- There are more villians than heros.
|
63 |
+
- The directions should be read.
|
64 |
+
- Mezarlar derin ormanlarda saklandı.
|
65 |
+
- source_sentence: K is a musician.
|
66 |
+
sentences:
|
67 |
+
- Klimt draws hotdogs.
|
68 |
+
- Ed Wood hiç mahkemeye çıkmadı.
|
69 |
+
- Çeçen Rusya yönetimi ele geçirdi.
|
70 |
+
- source_sentence: We moved closer.
|
71 |
+
sentences:
|
72 |
+
- Clinton is unaware of the process.
|
73 |
+
- Nesil deneyimleri anlamsızdır.
|
74 |
+
- Hormonların etkileri vardır.
|
75 |
+
pipeline_tag: sentence-similarity
|
76 |
+
model-index:
|
77 |
+
- name: SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
|
78 |
+
results:
|
79 |
+
- task:
|
80 |
+
type: semantic-similarity
|
81 |
+
name: Semantic Similarity
|
82 |
+
dataset:
|
83 |
+
name: tr ling
|
84 |
+
type: tr_ling
|
85 |
+
metrics:
|
86 |
+
- type: pearson_cosine
|
87 |
+
value: 0.058743115070889876
|
88 |
+
name: Pearson Cosine
|
89 |
+
- type: spearman_cosine
|
90 |
+
value: 0.059526247945378225
|
91 |
+
name: Spearman Cosine
|
92 |
+
- type: pearson_manhattan
|
93 |
+
value: 0.04582145815494953
|
94 |
+
name: Pearson Manhattan
|
95 |
+
- type: spearman_manhattan
|
96 |
+
value: 0.04331287037397966
|
97 |
+
name: Spearman Manhattan
|
98 |
+
- type: pearson_euclidean
|
99 |
+
value: 0.04709170917685587
|
100 |
+
name: Pearson Euclidean
|
101 |
+
- type: spearman_euclidean
|
102 |
+
value: 0.04407504959649961
|
103 |
+
name: Spearman Euclidean
|
104 |
+
- type: pearson_dot
|
105 |
+
value: 0.08477622619519222
|
106 |
+
name: Pearson Dot
|
107 |
+
- type: spearman_dot
|
108 |
+
value: 0.08243745050110735
|
109 |
+
name: Spearman Dot
|
110 |
+
- type: pearson_max
|
111 |
+
value: 0.08477622619519222
|
112 |
+
name: Pearson Max
|
113 |
+
- type: spearman_max
|
114 |
+
value: 0.08243745050110735
|
115 |
+
name: Spearman Max
|
116 |
+
---
|
117 |
+
|
118 |
+
# SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
|
119 |
+
|
120 |
+
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2) on the [MoritzLaurer/multilingual-nli-26lang-2mil7](https://huggingface.co/datasets/MoritzLaurer/multilingual-nli-26lang-2mil7) dataset. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
|
121 |
+
|
122 |
+
## Model Details
|
123 |
+
|
124 |
+
### Model Description
|
125 |
+
- **Model Type:** Sentence Transformer
|
126 |
+
- **Base model:** [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2) <!-- at revision bf3bf13ab40c3157080a7ab344c831b9ad18b5eb -->
|
127 |
+
- **Maximum Sequence Length:** 128 tokens
|
128 |
+
- **Output Dimensionality:** 384 tokens
|
129 |
+
- **Similarity Function:** Cosine Similarity
|
130 |
+
- **Training Dataset:**
|
131 |
+
- [MoritzLaurer/multilingual-nli-26lang-2mil7](https://huggingface.co/datasets/MoritzLaurer/multilingual-nli-26lang-2mil7)
|
132 |
+
- **Languages:** multilingual, zh, ja, ar, ko, de, fr, es, pt, hi, id, it, tr, ru, bn, ur, mr, ta, vi, fa, pl, uk, nl, sv, he, sw, ps
|
133 |
+
<!-- - **License:** Unknown -->
|
134 |
+
|
135 |
+
### Model Sources
|
136 |
+
|
137 |
+
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
|
138 |
+
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
|
139 |
+
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
|
140 |
+
|
141 |
+
### Full Model Architecture
|
142 |
+
|
143 |
+
```
|
144 |
+
SentenceTransformer(
|
145 |
+
(0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel
|
146 |
+
(1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
|
147 |
+
)
|
148 |
+
```
|
149 |
+
|
150 |
+
## Usage
|
151 |
+
|
152 |
+
### Direct Usage (Sentence Transformers)
|
153 |
+
|
154 |
+
First install the Sentence Transformers library:
|
155 |
+
|
156 |
+
```bash
|
157 |
+
pip install -U sentence-transformers
|
158 |
+
```
|
159 |
+
|
160 |
+
Then you can load this model and run inference.
|
161 |
+
```python
|
162 |
+
from sentence_transformers import SentenceTransformer
|
163 |
+
|
164 |
+
# Download from the 🤗 Hub
|
165 |
+
model = SentenceTransformer("sentence_transformers_model_id")
|
166 |
+
# Run inference
|
167 |
+
sentences = [
|
168 |
+
'We moved closer.',
|
169 |
+
'Clinton is unaware of the process.',
|
170 |
+
'Nesil deneyimleri anlamsızdır.',
|
171 |
+
]
|
172 |
+
embeddings = model.encode(sentences)
|
173 |
+
print(embeddings.shape)
|
174 |
+
# [3, 384]
|
175 |
+
|
176 |
+
# Get the similarity scores for the embeddings
|
177 |
+
similarities = model.similarity(embeddings, embeddings)
|
178 |
+
print(similarities.shape)
|
179 |
+
# [3, 3]
|
180 |
+
```
|
181 |
+
|
182 |
+
<!--
|
183 |
+
### Direct Usage (Transformers)
|
184 |
+
|
185 |
+
<details><summary>Click to see the direct usage in Transformers</summary>
|
186 |
+
|
187 |
+
</details>
|
188 |
+
-->
|
189 |
+
|
190 |
+
<!--
|
191 |
+
### Downstream Usage (Sentence Transformers)
|
192 |
+
|
193 |
+
You can finetune this model on your own dataset.
|
194 |
+
|
195 |
+
<details><summary>Click to expand</summary>
|
196 |
+
|
197 |
+
</details>
|
198 |
+
-->
|
199 |
+
|
200 |
+
<!--
|
201 |
+
### Out-of-Scope Use
|
202 |
+
|
203 |
+
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
|
204 |
+
-->
|
205 |
+
|
206 |
+
## Evaluation
|
207 |
+
|
208 |
+
### Metrics
|
209 |
+
|
210 |
+
#### Semantic Similarity
|
211 |
+
* Dataset: `tr_ling`
|
212 |
+
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
|
213 |
+
|
214 |
+
| Metric | Value |
|
215 |
+
|:-------------------|:-----------|
|
216 |
+
| pearson_cosine | 0.0587 |
|
217 |
+
| spearman_cosine | 0.0595 |
|
218 |
+
| pearson_manhattan | 0.0458 |
|
219 |
+
| spearman_manhattan | 0.0433 |
|
220 |
+
| pearson_euclidean | 0.0471 |
|
221 |
+
| spearman_euclidean | 0.0441 |
|
222 |
+
| pearson_dot | 0.0848 |
|
223 |
+
| spearman_dot | 0.0824 |
|
224 |
+
| pearson_max | 0.0848 |
|
225 |
+
| **spearman_max** | **0.0824** |
|
226 |
+
|
227 |
+
<!--
|
228 |
+
## Bias, Risks and Limitations
|
229 |
+
|
230 |
+
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
|
231 |
+
-->
|
232 |
+
|
233 |
+
<!--
|
234 |
+
### Recommendations
|
235 |
+
|
236 |
+
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
|
237 |
+
-->
|
238 |
+
|
239 |
+
## Training Details
|
240 |
+
|
241 |
+
### Training Dataset
|
242 |
+
|
243 |
+
#### MoritzLaurer/multilingual-nli-26lang-2mil7
|
244 |
+
|
245 |
+
* Dataset: [MoritzLaurer/multilingual-nli-26lang-2mil7](https://huggingface.co/datasets/MoritzLaurer/multilingual-nli-26lang-2mil7) at [510a233](https://huggingface.co/datasets/MoritzLaurer/multilingual-nli-26lang-2mil7/tree/510a233972a0d7ff0f767d82f46e046832c10538)
|
246 |
+
* Size: 25,000 training samples
|
247 |
+
* Columns: <code>premise_original</code>, <code>hypothesis_original</code>, <code>score</code>, <code>sentence1</code>, and <code>sentence2</code>
|
248 |
+
* Approximate statistics based on the first 1000 samples:
|
249 |
+
| | premise_original | hypothesis_original | score | sentence1 | sentence2 |
|
250 |
+
|:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
|
251 |
+
| type | string | string | int | string | string |
|
252 |
+
| details | <ul><li>min: 4 tokens</li><li>mean: 29.3 tokens</li><li>max: 107 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 15.62 tokens</li><li>max: 40 tokens</li></ul> | <ul><li>0: ~34.50%</li><li>1: ~33.30%</li><li>2: ~32.20%</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 28.28 tokens</li><li>max: 101 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 15.39 tokens</li><li>max: 38 tokens</li></ul> |
|
253 |
+
* Samples:
|
254 |
+
| premise_original | hypothesis_original | score | sentence1 | sentence2 |
|
255 |
+
|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------|
|
256 |
+
| <code>N, the total number of LC50 values used in calculating the CV(%) varied with organism and toxicant because some data were rejected due to water hardness, lack of concentration measurements, and/or because some of the LC50s were not calculable.</code> | <code>Most discarded data was rejected due to water hardness.</code> | <code>1</code> | <code>N, CV'nin hesaplanmasında kullanılan LC50 değerlerinin toplam sayısı (%) organizma ve toksik madde ile çeşitlidir, çünkü bazı veriler su sertliği, konsantrasyon ölçümlerinin eksikliği ve / veya LC50'lerin bazıları hesaplanamaz olduğu için reddedilmiştir.</code> | <code>Atılan verilerin çoğu su sertliği nedeniyle reddedildi.</code> |
|
257 |
+
| <code>As the home of the Venus de Milo and Mona Lisa, the Louvre drew almost unmanageable crowds until President Mitterrand ordered its re-organization in the 1980s.</code> | <code>The Louvre is home of the Venus de Milo and Mona Lisa.</code> | <code>0</code> | <code>Venus de Milo ve Mona Lisa'nın evi olarak Louvre, Başkan Mitterrand'ın 1980'lerde yeniden düzenlenmesini emredene kadar neredeyse yönetilemez kalabalıklar çekti.</code> | <code>Louvre, Venus de Milo ve Mona Lisa'nın evidir.</code> |
|
258 |
+
| <code>A year ago, the wife of the Oxford don noticed that the pattern on Kleenex quilted tissue uncannily resembled the Penrose Arrowed Rhombi tilings pattern, which Sir Roger had invented--and copyrighted--in 1974.</code> | <code>It has been recently found out a similarity between the pattern on the recent Kleenex quilted tissue and the one of the Penrose Arrowed Rhombi tilings.</code> | <code>0</code> | <code>Bir yıl önce Oxford'un karısı, Kleenex kapitone dokudaki desenin 1974'te Sir Roger'ın icat ettiği -ve telif hakkı olan - Penrose Arrowed Rhombi tilings desenine benzediğini fark etti.</code> | <code>Yakın zamanda, son Kleenex kapitone dokudaki desen ile Penrose Arrowed Rhombi döşemelerinden biri arasında bir benzerlik bulunmuştur.</code> |
|
259 |
+
* Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:
|
260 |
+
```json
|
261 |
+
{
|
262 |
+
"scale": 20.0,
|
263 |
+
"similarity_fct": "pairwise_cos_sim"
|
264 |
+
}
|
265 |
+
```
|
266 |
+
|
267 |
+
### Evaluation Dataset
|
268 |
+
|
269 |
+
#### MoritzLaurer/multilingual-nli-26lang-2mil7
|
270 |
+
|
271 |
+
* Dataset: [MoritzLaurer/multilingual-nli-26lang-2mil7](https://huggingface.co/datasets/MoritzLaurer/multilingual-nli-26lang-2mil7) at [510a233](https://huggingface.co/datasets/MoritzLaurer/multilingual-nli-26lang-2mil7/tree/510a233972a0d7ff0f767d82f46e046832c10538)
|
272 |
+
* Size: 5,000 evaluation samples
|
273 |
+
* Columns: <code>premise_original</code>, <code>hypothesis_original</code>, <code>score</code>, <code>sentence1</code>, and <code>sentence2</code>
|
274 |
+
* Approximate statistics based on the first 1000 samples:
|
275 |
+
| | premise_original | hypothesis_original | score | sentence1 | sentence2 |
|
276 |
+
|:--------|:---------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
|
277 |
+
| type | string | string | int | string | string |
|
278 |
+
| details | <ul><li>min: 5 tokens</li><li>mean: 30.3 tokens</li><li>max: 99 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 15.11 tokens</li><li>max: 56 tokens</li></ul> | <ul><li>0: ~34.50%</li><li>1: ~29.90%</li><li>2: ~35.60%</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 29.94 tokens</li><li>max: 106 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 15.29 tokens</li><li>max: 52 tokens</li></ul> |
|
279 |
+
* Samples:
|
280 |
+
| premise_original | hypothesis_original | score | sentence1 | sentence2 |
|
281 |
+
|:----------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------|:---------------|:------------------------------------------------------------------------------|:-----------------------------------------------------------------|
|
282 |
+
| <code>But the racism charge isn't quirky or wacky--it's demagogy.</code> | <code>The accusation of prejudice based on a pedestrian kind of hatred.</code> | <code>0</code> | <code>Ama ırkçılık suçlaması tuhaf ya da tuhaf değil, bu bir demagoji.</code> | <code>Yaya nefretine dayanan önyargı suçlaması.</code> |
|
283 |
+
| <code>Why would Gates allow the publication of such a book with his byline and photo on the dust jacket?</code> | <code>Gates' byline and photo are on the dust jacket</code> | <code>0</code> | <code>Gates neden böyle bir kitabın basılmasına izin versin ki?</code> | <code>Gates'in çizgisi ve fotoğrafı toz ceketin üzerinde.</code> |
|
284 |
+
| <code>I am a nonsmoker and allergic to cigarette smoke.</code> | <code>I do not smoke.</code> | <code>0</code> | <code>Sigara içmeyen biriyim ve sigara dumanına alerjim var.</code> | <code>Sigara içmiyorum.</code> |
|
285 |
+
* Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:
|
286 |
+
```json
|
287 |
+
{
|
288 |
+
"scale": 20.0,
|
289 |
+
"similarity_fct": "pairwise_cos_sim"
|
290 |
+
}
|
291 |
+
```
|
292 |
+
|
293 |
+
### Training Hyperparameters
|
294 |
+
#### Non-Default Hyperparameters
|
295 |
+
|
296 |
+
- `eval_strategy`: epoch
|
297 |
+
- `per_device_train_batch_size`: 32
|
298 |
+
- `per_device_eval_batch_size`: 64
|
299 |
+
- `learning_rate`: 2e-05
|
300 |
+
- `num_train_epochs`: 5
|
301 |
+
- `warmup_ratio`: 0.1
|
302 |
+
- `fp16`: True
|
303 |
+
- `load_best_model_at_end`: True
|
304 |
+
- `ddp_find_unused_parameters`: False
|
305 |
+
|
306 |
+
#### All Hyperparameters
|
307 |
+
<details><summary>Click to expand</summary>
|
308 |
+
|
309 |
+
- `overwrite_output_dir`: False
|
310 |
+
- `do_predict`: False
|
311 |
+
- `eval_strategy`: epoch
|
312 |
+
- `prediction_loss_only`: True
|
313 |
+
- `per_device_train_batch_size`: 32
|
314 |
+
- `per_device_eval_batch_size`: 64
|
315 |
+
- `per_gpu_train_batch_size`: None
|
316 |
+
- `per_gpu_eval_batch_size`: None
|
317 |
+
- `gradient_accumulation_steps`: 1
|
318 |
+
- `eval_accumulation_steps`: None
|
319 |
+
- `learning_rate`: 2e-05
|
320 |
+
- `weight_decay`: 0.0
|
321 |
+
- `adam_beta1`: 0.9
|
322 |
+
- `adam_beta2`: 0.999
|
323 |
+
- `adam_epsilon`: 1e-08
|
324 |
+
- `max_grad_norm`: 1.0
|
325 |
+
- `num_train_epochs`: 5
|
326 |
+
- `max_steps`: -1
|
327 |
+
- `lr_scheduler_type`: linear
|
328 |
+
- `lr_scheduler_kwargs`: {}
|
329 |
+
- `warmup_ratio`: 0.1
|
330 |
+
- `warmup_steps`: 0
|
331 |
+
- `log_level`: passive
|
332 |
+
- `log_level_replica`: warning
|
333 |
+
- `log_on_each_node`: True
|
334 |
+
- `logging_nan_inf_filter`: True
|
335 |
+
- `save_safetensors`: True
|
336 |
+
- `save_on_each_node`: False
|
337 |
+
- `save_only_model`: False
|
338 |
+
- `restore_callback_states_from_checkpoint`: False
|
339 |
+
- `no_cuda`: False
|
340 |
+
- `use_cpu`: False
|
341 |
+
- `use_mps_device`: False
|
342 |
+
- `seed`: 42
|
343 |
+
- `data_seed`: None
|
344 |
+
- `jit_mode_eval`: False
|
345 |
+
- `use_ipex`: False
|
346 |
+
- `bf16`: False
|
347 |
+
- `fp16`: True
|
348 |
+
- `fp16_opt_level`: O1
|
349 |
+
- `half_precision_backend`: auto
|
350 |
+
- `bf16_full_eval`: False
|
351 |
+
- `fp16_full_eval`: False
|
352 |
+
- `tf32`: None
|
353 |
+
- `local_rank`: 0
|
354 |
+
- `ddp_backend`: None
|
355 |
+
- `tpu_num_cores`: None
|
356 |
+
- `tpu_metrics_debug`: False
|
357 |
+
- `debug`: []
|
358 |
+
- `dataloader_drop_last`: False
|
359 |
+
- `dataloader_num_workers`: 0
|
360 |
+
- `dataloader_prefetch_factor`: None
|
361 |
+
- `past_index`: -1
|
362 |
+
- `disable_tqdm`: False
|
363 |
+
- `remove_unused_columns`: True
|
364 |
+
- `label_names`: None
|
365 |
+
- `load_best_model_at_end`: True
|
366 |
+
- `ignore_data_skip`: False
|
367 |
+
- `fsdp`: []
|
368 |
+
- `fsdp_min_num_params`: 0
|
369 |
+
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
|
370 |
+
- `fsdp_transformer_layer_cls_to_wrap`: None
|
371 |
+
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
|
372 |
+
- `deepspeed`: None
|
373 |
+
- `label_smoothing_factor`: 0.0
|
374 |
+
- `optim`: adamw_torch
|
375 |
+
- `optim_args`: None
|
376 |
+
- `adafactor`: False
|
377 |
+
- `group_by_length`: False
|
378 |
+
- `length_column_name`: length
|
379 |
+
- `ddp_find_unused_parameters`: False
|
380 |
+
- `ddp_bucket_cap_mb`: None
|
381 |
+
- `ddp_broadcast_buffers`: False
|
382 |
+
- `dataloader_pin_memory`: True
|
383 |
+
- `dataloader_persistent_workers`: False
|
384 |
+
- `skip_memory_metrics`: True
|
385 |
+
- `use_legacy_prediction_loop`: False
|
386 |
+
- `push_to_hub`: False
|
387 |
+
- `resume_from_checkpoint`: None
|
388 |
+
- `hub_model_id`: None
|
389 |
+
- `hub_strategy`: every_save
|
390 |
+
- `hub_private_repo`: False
|
391 |
+
- `hub_always_push`: False
|
392 |
+
- `gradient_checkpointing`: False
|
393 |
+
- `gradient_checkpointing_kwargs`: None
|
394 |
+
- `include_inputs_for_metrics`: False
|
395 |
+
- `eval_do_concat_batches`: True
|
396 |
+
- `fp16_backend`: auto
|
397 |
+
- `push_to_hub_model_id`: None
|
398 |
+
- `push_to_hub_organization`: None
|
399 |
+
- `mp_parameters`:
|
400 |
+
- `auto_find_batch_size`: False
|
401 |
+
- `full_determinism`: False
|
402 |
+
- `torchdynamo`: None
|
403 |
+
- `ray_scope`: last
|
404 |
+
- `ddp_timeout`: 1800
|
405 |
+
- `torch_compile`: False
|
406 |
+
- `torch_compile_backend`: None
|
407 |
+
- `torch_compile_mode`: None
|
408 |
+
- `dispatch_batches`: None
|
409 |
+
- `split_batches`: None
|
410 |
+
- `include_tokens_per_second`: False
|
411 |
+
- `include_num_input_tokens_seen`: False
|
412 |
+
- `neftune_noise_alpha`: None
|
413 |
+
- `optim_target_modules`: None
|
414 |
+
- `batch_eval_metrics`: False
|
415 |
+
- `batch_sampler`: batch_sampler
|
416 |
+
- `multi_dataset_batch_sampler`: proportional
|
417 |
+
|
418 |
+
</details>
|
419 |
+
|
420 |
+
### Training Logs
|
421 |
+
<details><summary>Click to expand</summary>
|
422 |
+
|
423 |
+
| Epoch | Step | Training Loss | loss | tr_ling_spearman_max |
|
424 |
+
|:------:|:----:|:-------------:|:------:|:--------------------:|
|
425 |
+
| 0.0320 | 25 | 17.17 | - | - |
|
426 |
+
| 0.0639 | 50 | 16.4932 | - | - |
|
427 |
+
| 0.0959 | 75 | 16.5976 | - | - |
|
428 |
+
| 0.1279 | 100 | 15.6991 | - | - |
|
429 |
+
| 0.1598 | 125 | 14.876 | - | - |
|
430 |
+
| 0.1918 | 150 | 14.4828 | - | - |
|
431 |
+
| 0.2238 | 175 | 12.7061 | - | - |
|
432 |
+
| 0.2558 | 200 | 10.8687 | - | - |
|
433 |
+
| 0.2877 | 225 | 8.3797 | - | - |
|
434 |
+
| 0.3197 | 250 | 6.2029 | - | - |
|
435 |
+
| 0.3517 | 275 | 5.8228 | - | - |
|
436 |
+
| 0.3836 | 300 | 5.811 | - | - |
|
437 |
+
| 0.4156 | 325 | 5.8079 | - | - |
|
438 |
+
| 0.4476 | 350 | 5.8077 | - | - |
|
439 |
+
| 0.4795 | 375 | 5.8035 | - | - |
|
440 |
+
| 0.5115 | 400 | 5.8072 | - | - |
|
441 |
+
| 0.5435 | 425 | 5.8033 | - | - |
|
442 |
+
| 0.5754 | 450 | 5.8086 | - | - |
|
443 |
+
| 0.6074 | 475 | 5.81 | - | - |
|
444 |
+
| 0.6394 | 500 | 5.7949 | - | - |
|
445 |
+
| 0.6714 | 525 | 5.8079 | - | - |
|
446 |
+
| 0.7033 | 550 | 5.8057 | - | - |
|
447 |
+
| 0.7353 | 575 | 5.8097 | - | - |
|
448 |
+
| 0.7673 | 600 | 5.7986 | - | - |
|
449 |
+
| 0.7992 | 625 | 5.8051 | - | - |
|
450 |
+
| 0.8312 | 650 | 5.8041 | - | - |
|
451 |
+
| 0.8632 | 675 | 5.7907 | - | - |
|
452 |
+
| 0.8951 | 700 | 5.7991 | - | - |
|
453 |
+
| 0.9271 | 725 | 5.8035 | - | - |
|
454 |
+
| 0.9591 | 750 | 5.7945 | - | - |
|
455 |
+
| 0.9910 | 775 | 5.8077 | - | - |
|
456 |
+
| 1.0 | 782 | - | 5.8024 | 0.0330 |
|
457 |
+
| 1.0230 | 800 | 5.6703 | - | - |
|
458 |
+
| 1.0550 | 825 | 5.8052 | - | - |
|
459 |
+
| 1.0870 | 850 | 5.7936 | - | - |
|
460 |
+
| 1.1189 | 875 | 5.7924 | - | - |
|
461 |
+
| 1.1509 | 900 | 5.7806 | - | - |
|
462 |
+
| 1.1829 | 925 | 5.7835 | - | - |
|
463 |
+
| 1.2148 | 950 | 5.7619 | - | - |
|
464 |
+
| 1.2468 | 975 | 5.8038 | - | - |
|
465 |
+
| 1.2788 | 1000 | 5.779 | - | - |
|
466 |
+
| 1.3107 | 1025 | 5.7904 | - | - |
|
467 |
+
| 1.3427 | 1050 | 5.7696 | - | - |
|
468 |
+
| 1.3747 | 1075 | 5.7919 | - | - |
|
469 |
+
| 1.4066 | 1100 | 5.7785 | - | - |
|
470 |
+
| 1.4386 | 1125 | 5.7862 | - | - |
|
471 |
+
| 1.4706 | 1150 | 5.7703 | - | - |
|
472 |
+
| 1.5026 | 1175 | 5.773 | - | - |
|
473 |
+
| 1.5345 | 1200 | 5.7627 | - | - |
|
474 |
+
| 1.5665 | 1225 | 5.7596 | - | - |
|
475 |
+
| 1.5985 | 1250 | 5.7882 | - | - |
|
476 |
+
| 1.6304 | 1275 | 5.7828 | - | - |
|
477 |
+
| 1.6624 | 1300 | 5.771 | - | - |
|
478 |
+
| 1.6944 | 1325 | 5.788 | - | - |
|
479 |
+
| 1.7263 | 1350 | 5.7719 | - | - |
|
480 |
+
| 1.7583 | 1375 | 5.7846 | - | - |
|
481 |
+
| 1.7903 | 1400 | 5.7838 | - | - |
|
482 |
+
| 1.8223 | 1425 | 5.7912 | - | - |
|
483 |
+
| 1.8542 | 1450 | 5.7686 | - | - |
|
484 |
+
| 1.8862 | 1475 | 5.7938 | - | - |
|
485 |
+
| 1.9182 | 1500 | 5.7847 | - | - |
|
486 |
+
| 1.9501 | 1525 | 5.7952 | - | - |
|
487 |
+
| 1.9821 | 1550 | 5.7528 | - | - |
|
488 |
+
| 2.0 | 1564 | - | 5.7933 | 0.0682 |
|
489 |
+
| 2.0141 | 1575 | 5.65 | - | - |
|
490 |
+
| 2.0460 | 1600 | 5.7537 | - | - |
|
491 |
+
| 2.0780 | 1625 | 5.7098 | - | - |
|
492 |
+
| 2.1100 | 1650 | 5.7149 | - | - |
|
493 |
+
| 2.1419 | 1675 | 5.7585 | - | - |
|
494 |
+
| 2.1739 | 1700 | 5.7277 | - | - |
|
495 |
+
| 2.2059 | 1725 | 5.7482 | - | - |
|
496 |
+
| 2.2379 | 1750 | 5.7115 | - | - |
|
497 |
+
| 2.2698 | 1775 | 5.6895 | - | - |
|
498 |
+
| 2.3018 | 1800 | 5.7389 | - | - |
|
499 |
+
| 2.3338 | 1825 | 5.7161 | - | - |
|
500 |
+
| 2.3657 | 1850 | 5.7123 | - | - |
|
501 |
+
| 2.3977 | 1875 | 5.7322 | - | - |
|
502 |
+
| 2.4297 | 1900 | 5.7421 | - | - |
|
503 |
+
| 2.4616 | 1925 | 5.7615 | - | - |
|
504 |
+
| 2.4936 | 1950 | 5.7493 | - | - |
|
505 |
+
| 2.5256 | 1975 | 5.7298 | - | - |
|
506 |
+
| 2.5575 | 2000 | 5.7529 | - | - |
|
507 |
+
| 2.5895 | 2025 | 5.7318 | - | - |
|
508 |
+
| 2.6215 | 2050 | 5.7036 | - | - |
|
509 |
+
| 2.6535 | 2075 | 5.7158 | - | - |
|
510 |
+
| 2.6854 | 2100 | 5.7209 | - | - |
|
511 |
+
| 2.7174 | 2125 | 5.738 | - | - |
|
512 |
+
| 2.7494 | 2150 | 5.7337 | - | - |
|
513 |
+
| 2.7813 | 2175 | 5.713 | - | - |
|
514 |
+
| 2.8133 | 2200 | 5.7257 | - | - |
|
515 |
+
| 2.8453 | 2225 | 5.6958 | - | - |
|
516 |
+
| 2.8772 | 2250 | 5.7053 | - | - |
|
517 |
+
| 2.9092 | 2275 | 5.7246 | - | - |
|
518 |
+
| 2.9412 | 2300 | 5.7291 | - | - |
|
519 |
+
| 2.9731 | 2325 | 5.7139 | - | - |
|
520 |
+
| 3.0 | 2346 | - | 5.8510 | 0.0837 |
|
521 |
+
| 3.0051 | 2350 | 5.5715 | - | - |
|
522 |
+
| 3.0371 | 2375 | 5.6558 | - | - |
|
523 |
+
| 3.0691 | 2400 | 5.6441 | - | - |
|
524 |
+
| 3.1010 | 2425 | 5.6569 | - | - |
|
525 |
+
| 3.1330 | 2450 | 5.669 | - | - |
|
526 |
+
| 3.1650 | 2475 | 5.6361 | - | - |
|
527 |
+
| 3.1969 | 2500 | 5.6524 | - | - |
|
528 |
+
| 3.2289 | 2525 | 5.6773 | - | - |
|
529 |
+
| 3.2609 | 2550 | 5.6552 | - | - |
|
530 |
+
| 3.2928 | 2575 | 5.6807 | - | - |
|
531 |
+
| 3.3248 | 2600 | 5.6638 | - | - |
|
532 |
+
| 3.3568 | 2625 | 5.6582 | - | - |
|
533 |
+
| 3.3887 | 2650 | 5.658 | - | - |
|
534 |
+
| 3.4207 | 2675 | 5.6626 | - | - |
|
535 |
+
| 3.4527 | 2700 | 5.6802 | - | - |
|
536 |
+
| 3.4847 | 2725 | 5.6377 | - | - |
|
537 |
+
| 3.5166 | 2750 | 5.6752 | - | - |
|
538 |
+
| 3.5486 | 2775 | 5.6573 | - | - |
|
539 |
+
| 3.5806 | 2800 | 5.6963 | - | - |
|
540 |
+
| 3.6125 | 2825 | 5.7007 | - | - |
|
541 |
+
| 3.6445 | 2850 | 5.6746 | - | - |
|
542 |
+
| 3.6765 | 2875 | 5.6312 | - | - |
|
543 |
+
| 3.7084 | 2900 | 5.5596 | - | - |
|
544 |
+
| 3.7404 | 2925 | 5.7003 | - | - |
|
545 |
+
| 3.7724 | 2950 | 5.6739 | - | - |
|
546 |
+
| 3.8043 | 2975 | 5.655 | - | - |
|
547 |
+
| 3.8363 | 3000 | 5.6787 | - | - |
|
548 |
+
| 3.8683 | 3025 | 5.643 | - | - |
|
549 |
+
| 3.9003 | 3050 | 5.6412 | - | - |
|
550 |
+
| 3.9322 | 3075 | 5.758 | - | - |
|
551 |
+
| 3.9642 | 3100 | 5.6769 | - | - |
|
552 |
+
| 3.9962 | 3125 | 5.7206 | - | - |
|
553 |
+
| 4.0 | 3128 | - | 5.9125 | 0.0824 |
|
554 |
+
|
555 |
+
</details>
|
556 |
+
|
557 |
+
### Framework Versions
|
558 |
+
- Python: 3.10.12
|
559 |
+
- Sentence Transformers: 3.0.0
|
560 |
+
- Transformers: 4.41.0
|
561 |
+
- PyTorch: 2.3.0+cu121
|
562 |
+
- Accelerate: 0.30.1
|
563 |
+
- Datasets: 2.19.1
|
564 |
+
- Tokenizers: 0.19.1
|
565 |
+
|
566 |
+
## Citation
|
567 |
+
|
568 |
+
### BibTeX
|
569 |
+
|
570 |
+
#### Sentence Transformers
|
571 |
+
```bibtex
|
572 |
+
@inproceedings{reimers-2019-sentence-bert,
|
573 |
+
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
|
574 |
+
author = "Reimers, Nils and Gurevych, Iryna",
|
575 |
+
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
|
576 |
+
month = "11",
|
577 |
+
year = "2019",
|
578 |
+
publisher = "Association for Computational Linguistics",
|
579 |
+
url = "https://arxiv.org/abs/1908.10084",
|
580 |
+
}
|
581 |
+
```
|
582 |
+
|
583 |
+
#### CoSENTLoss
|
584 |
+
```bibtex
|
585 |
+
@online{kexuefm-8847,
|
586 |
+
title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
|
587 |
+
author={Su Jianlin},
|
588 |
+
year={2022},
|
589 |
+
month={Jan},
|
590 |
+
url={https://kexue.fm/archives/8847},
|
591 |
+
}
|
592 |
+
```
|
593 |
+
|
594 |
+
<!--
|
595 |
+
## Glossary
|
596 |
+
|
597 |
+
*Clearly define terms in order to be accessible across audiences.*
|
598 |
+
-->
|
599 |
+
|
600 |
+
<!--
|
601 |
+
## Model Card Authors
|
602 |
+
|
603 |
+
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
|
604 |
+
-->
|
605 |
+
|
606 |
+
<!--
|
607 |
+
## Model Card Contact
|
608 |
+
|
609 |
+
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
|
610 |
+
-->
|
checkpoint-3128/config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2",
|
3 |
+
"architectures": [
|
4 |
+
"BertModel"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"classifier_dropout": null,
|
8 |
+
"gradient_checkpointing": false,
|
9 |
+
"hidden_act": "gelu",
|
10 |
+
"hidden_dropout_prob": 0.1,
|
11 |
+
"hidden_size": 384,
|
12 |
+
"initializer_range": 0.02,
|
13 |
+
"intermediate_size": 1536,
|
14 |
+
"layer_norm_eps": 1e-12,
|
15 |
+
"max_position_embeddings": 512,
|
16 |
+
"model_type": "bert",
|
17 |
+
"num_attention_heads": 12,
|
18 |
+
"num_hidden_layers": 12,
|
19 |
+
"pad_token_id": 0,
|
20 |
+
"position_embedding_type": "absolute",
|
21 |
+
"torch_dtype": "float32",
|
22 |
+
"transformers_version": "4.41.0",
|
23 |
+
"type_vocab_size": 2,
|
24 |
+
"use_cache": true,
|
25 |
+
"vocab_size": 250037
|
26 |
+
}
|
checkpoint-3128/config_sentence_transformers.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"__version__": {
|
3 |
+
"sentence_transformers": "2.0.0",
|
4 |
+
"transformers": "4.7.0",
|
5 |
+
"pytorch": "1.9.0+cu102"
|
6 |
+
},
|
7 |
+
"prompts": {},
|
8 |
+
"default_prompt_name": null,
|
9 |
+
"similarity_fn_name": null
|
10 |
+
}
|
checkpoint-3128/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:19f15c426e0a8e3ede12b4f3bb95161ed5a4c5a51bcb20ab519919596e330412
|
3 |
+
size 470637416
|
checkpoint-3128/modules.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
[
|
2 |
+
{
|
3 |
+
"idx": 0,
|
4 |
+
"name": "0",
|
5 |
+
"path": "",
|
6 |
+
"type": "sentence_transformers.models.Transformer"
|
7 |
+
},
|
8 |
+
{
|
9 |
+
"idx": 1,
|
10 |
+
"name": "1",
|
11 |
+
"path": "1_Pooling",
|
12 |
+
"type": "sentence_transformers.models.Pooling"
|
13 |
+
}
|
14 |
+
]
|
checkpoint-3128/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5a405964bc0e82a7e1469514b996b71e65018ff76733fce47d9f2788f24fc5ab
|
3 |
+
size 940212218
|
checkpoint-3128/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:46c53b1054e154626ecf95cb41fffe8108cccf21bc4843652c954100009a36a3
|
3 |
+
size 14180
|
checkpoint-3128/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d72a55fbec173081120a6cdf969c078f1cc9ef4f2e0dc0f64b604e716642a81e
|
3 |
+
size 1064
|
checkpoint-3128/sentence_bert_config.json
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"max_seq_length": 128,
|
3 |
+
"do_lower_case": false
|
4 |
+
}
|
checkpoint-3128/special_tokens_map.json
ADDED
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<s>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"cls_token": {
|
10 |
+
"content": "<s>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"eos_token": {
|
17 |
+
"content": "</s>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
},
|
23 |
+
"mask_token": {
|
24 |
+
"content": "<mask>",
|
25 |
+
"lstrip": true,
|
26 |
+
"normalized": false,
|
27 |
+
"rstrip": false,
|
28 |
+
"single_word": false
|
29 |
+
},
|
30 |
+
"pad_token": {
|
31 |
+
"content": "<pad>",
|
32 |
+
"lstrip": false,
|
33 |
+
"normalized": false,
|
34 |
+
"rstrip": false,
|
35 |
+
"single_word": false
|
36 |
+
},
|
37 |
+
"sep_token": {
|
38 |
+
"content": "</s>",
|
39 |
+
"lstrip": false,
|
40 |
+
"normalized": false,
|
41 |
+
"rstrip": false,
|
42 |
+
"single_word": false
|
43 |
+
},
|
44 |
+
"unk_token": {
|
45 |
+
"content": "<unk>",
|
46 |
+
"lstrip": false,
|
47 |
+
"normalized": false,
|
48 |
+
"rstrip": false,
|
49 |
+
"single_word": false
|
50 |
+
}
|
51 |
+
}
|
checkpoint-3128/tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cad551d5600a84242d0973327029452a1e3672ba6313c2a3c3d69c4310e12719
|
3 |
+
size 17082987
|
checkpoint-3128/tokenizer_config.json
ADDED
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "<s>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "<pad>",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "</s>",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"3": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"250001": {
|
36 |
+
"content": "<mask>",
|
37 |
+
"lstrip": true,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
}
|
43 |
+
},
|
44 |
+
"bos_token": "<s>",
|
45 |
+
"clean_up_tokenization_spaces": true,
|
46 |
+
"cls_token": "<s>",
|
47 |
+
"do_lower_case": true,
|
48 |
+
"eos_token": "</s>",
|
49 |
+
"mask_token": "<mask>",
|
50 |
+
"max_length": 128,
|
51 |
+
"model_max_length": 128,
|
52 |
+
"pad_to_multiple_of": null,
|
53 |
+
"pad_token": "<pad>",
|
54 |
+
"pad_token_type_id": 0,
|
55 |
+
"padding_side": "right",
|
56 |
+
"sep_token": "</s>",
|
57 |
+
"stride": 0,
|
58 |
+
"strip_accents": null,
|
59 |
+
"tokenize_chinese_chars": true,
|
60 |
+
"tokenizer_class": "BertTokenizer",
|
61 |
+
"truncation_side": "right",
|
62 |
+
"truncation_strategy": "longest_first",
|
63 |
+
"unk_token": "<unk>"
|
64 |
+
}
|
checkpoint-3128/trainer_state.json
ADDED
@@ -0,0 +1,989 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 5.793323516845703,
|
3 |
+
"best_model_checkpoint": "turkish-embedding-model/checkpoint-1564",
|
4 |
+
"epoch": 4.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 3128,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0319693094629156,
|
13 |
+
"grad_norm": 70.72708129882812,
|
14 |
+
"learning_rate": 1.1253196930946293e-06,
|
15 |
+
"loss": 17.17,
|
16 |
+
"step": 25
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.0639386189258312,
|
20 |
+
"grad_norm": 81.68770599365234,
|
21 |
+
"learning_rate": 2.4040920716112534e-06,
|
22 |
+
"loss": 16.4932,
|
23 |
+
"step": 50
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.0959079283887468,
|
27 |
+
"grad_norm": 109.91338348388672,
|
28 |
+
"learning_rate": 3.6828644501278778e-06,
|
29 |
+
"loss": 16.5976,
|
30 |
+
"step": 75
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.1278772378516624,
|
34 |
+
"grad_norm": 73.892578125,
|
35 |
+
"learning_rate": 4.961636828644502e-06,
|
36 |
+
"loss": 15.6991,
|
37 |
+
"step": 100
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.159846547314578,
|
41 |
+
"grad_norm": 79.35150909423828,
|
42 |
+
"learning_rate": 6.240409207161126e-06,
|
43 |
+
"loss": 14.876,
|
44 |
+
"step": 125
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.1918158567774936,
|
48 |
+
"grad_norm": 83.0904541015625,
|
49 |
+
"learning_rate": 7.5191815856777495e-06,
|
50 |
+
"loss": 14.4828,
|
51 |
+
"step": 150
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.2237851662404092,
|
55 |
+
"grad_norm": 76.82855987548828,
|
56 |
+
"learning_rate": 8.797953964194374e-06,
|
57 |
+
"loss": 12.7061,
|
58 |
+
"step": 175
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.2557544757033248,
|
62 |
+
"grad_norm": 51.30181121826172,
|
63 |
+
"learning_rate": 1.0076726342710998e-05,
|
64 |
+
"loss": 10.8687,
|
65 |
+
"step": 200
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.2877237851662404,
|
69 |
+
"grad_norm": 18.70808219909668,
|
70 |
+
"learning_rate": 1.1355498721227622e-05,
|
71 |
+
"loss": 8.3797,
|
72 |
+
"step": 225
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.319693094629156,
|
76 |
+
"grad_norm": 1.3039417266845703,
|
77 |
+
"learning_rate": 1.2634271099744246e-05,
|
78 |
+
"loss": 6.2029,
|
79 |
+
"step": 250
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.3516624040920716,
|
83 |
+
"grad_norm": 0.2324853092432022,
|
84 |
+
"learning_rate": 1.391304347826087e-05,
|
85 |
+
"loss": 5.8228,
|
86 |
+
"step": 275
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.3836317135549872,
|
90 |
+
"grad_norm": 0.1757364720106125,
|
91 |
+
"learning_rate": 1.5191815856777494e-05,
|
92 |
+
"loss": 5.811,
|
93 |
+
"step": 300
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.4156010230179028,
|
97 |
+
"grad_norm": 0.1788654774427414,
|
98 |
+
"learning_rate": 1.647058823529412e-05,
|
99 |
+
"loss": 5.8079,
|
100 |
+
"step": 325
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.4475703324808184,
|
104 |
+
"grad_norm": 0.12862567603588104,
|
105 |
+
"learning_rate": 1.7749360613810744e-05,
|
106 |
+
"loss": 5.8077,
|
107 |
+
"step": 350
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.479539641943734,
|
111 |
+
"grad_norm": 0.14497514069080353,
|
112 |
+
"learning_rate": 1.9028132992327367e-05,
|
113 |
+
"loss": 5.8035,
|
114 |
+
"step": 375
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.5115089514066496,
|
118 |
+
"grad_norm": 0.1350390762090683,
|
119 |
+
"learning_rate": 1.996589940323956e-05,
|
120 |
+
"loss": 5.8072,
|
121 |
+
"step": 400
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.5434782608695652,
|
125 |
+
"grad_norm": 0.1435602754354477,
|
126 |
+
"learning_rate": 1.9823813583404378e-05,
|
127 |
+
"loss": 5.8033,
|
128 |
+
"step": 425
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.5754475703324808,
|
132 |
+
"grad_norm": 0.11389254033565521,
|
133 |
+
"learning_rate": 1.96817277635692e-05,
|
134 |
+
"loss": 5.8086,
|
135 |
+
"step": 450
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.6074168797953964,
|
139 |
+
"grad_norm": 0.15821650624275208,
|
140 |
+
"learning_rate": 1.9539641943734017e-05,
|
141 |
+
"loss": 5.81,
|
142 |
+
"step": 475
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.639386189258312,
|
146 |
+
"grad_norm": 0.1179889366030693,
|
147 |
+
"learning_rate": 1.9397556123898838e-05,
|
148 |
+
"loss": 5.7949,
|
149 |
+
"step": 500
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.6713554987212276,
|
153 |
+
"grad_norm": 0.10912967473268509,
|
154 |
+
"learning_rate": 1.9255470304063656e-05,
|
155 |
+
"loss": 5.8079,
|
156 |
+
"step": 525
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.7033248081841432,
|
160 |
+
"grad_norm": 0.11702870577573776,
|
161 |
+
"learning_rate": 1.9113384484228477e-05,
|
162 |
+
"loss": 5.8057,
|
163 |
+
"step": 550
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.7352941176470589,
|
167 |
+
"grad_norm": 0.13132448494434357,
|
168 |
+
"learning_rate": 1.8971298664393295e-05,
|
169 |
+
"loss": 5.8097,
|
170 |
+
"step": 575
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.7672634271099744,
|
174 |
+
"grad_norm": 0.15833145380020142,
|
175 |
+
"learning_rate": 1.8829212844558116e-05,
|
176 |
+
"loss": 5.7986,
|
177 |
+
"step": 600
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.7992327365728901,
|
181 |
+
"grad_norm": 0.11651863902807236,
|
182 |
+
"learning_rate": 1.8687127024722937e-05,
|
183 |
+
"loss": 5.8051,
|
184 |
+
"step": 625
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.8312020460358056,
|
188 |
+
"grad_norm": 0.5393890142440796,
|
189 |
+
"learning_rate": 1.854504120488775e-05,
|
190 |
+
"loss": 5.8041,
|
191 |
+
"step": 650
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.8631713554987213,
|
195 |
+
"grad_norm": 0.6457561254501343,
|
196 |
+
"learning_rate": 1.8402955385052572e-05,
|
197 |
+
"loss": 5.7907,
|
198 |
+
"step": 675
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.8951406649616368,
|
202 |
+
"grad_norm": 0.5643135905265808,
|
203 |
+
"learning_rate": 1.8260869565217393e-05,
|
204 |
+
"loss": 5.7991,
|
205 |
+
"step": 700
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.9271099744245525,
|
209 |
+
"grad_norm": 3.214787721633911,
|
210 |
+
"learning_rate": 1.811878374538221e-05,
|
211 |
+
"loss": 5.8035,
|
212 |
+
"step": 725
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.959079283887468,
|
216 |
+
"grad_norm": 2.781162977218628,
|
217 |
+
"learning_rate": 1.7976697925547032e-05,
|
218 |
+
"loss": 5.7945,
|
219 |
+
"step": 750
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.9910485933503836,
|
223 |
+
"grad_norm": 0.38559335470199585,
|
224 |
+
"learning_rate": 1.783461210571185e-05,
|
225 |
+
"loss": 5.8077,
|
226 |
+
"step": 775
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 1.0,
|
230 |
+
"eval_loss": 5.8023600578308105,
|
231 |
+
"eval_runtime": 18.0632,
|
232 |
+
"eval_samples_per_second": 276.805,
|
233 |
+
"eval_steps_per_second": 4.374,
|
234 |
+
"eval_tr_ling_pearson_cosine": 0.017751548525136808,
|
235 |
+
"eval_tr_ling_pearson_dot": 0.025703597820631346,
|
236 |
+
"eval_tr_ling_pearson_euclidean": 0.02195284877201089,
|
237 |
+
"eval_tr_ling_pearson_manhattan": 0.02083376479528459,
|
238 |
+
"eval_tr_ling_pearson_max": 0.025703597820631346,
|
239 |
+
"eval_tr_ling_spearman_cosine": 0.027108099994157316,
|
240 |
+
"eval_tr_ling_spearman_dot": 0.03304394653738539,
|
241 |
+
"eval_tr_ling_spearman_euclidean": 0.025485959636772793,
|
242 |
+
"eval_tr_ling_spearman_manhattan": 0.024466610177699702,
|
243 |
+
"eval_tr_ling_spearman_max": 0.03304394653738539,
|
244 |
+
"step": 782
|
245 |
+
},
|
246 |
+
{
|
247 |
+
"epoch": 1.0230179028132993,
|
248 |
+
"grad_norm": 0.3645063638687134,
|
249 |
+
"learning_rate": 1.769252628587667e-05,
|
250 |
+
"loss": 5.6703,
|
251 |
+
"step": 800
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 1.054987212276215,
|
255 |
+
"grad_norm": 0.9638137817382812,
|
256 |
+
"learning_rate": 1.7550440466041488e-05,
|
257 |
+
"loss": 5.8052,
|
258 |
+
"step": 825
|
259 |
+
},
|
260 |
+
{
|
261 |
+
"epoch": 1.0869565217391304,
|
262 |
+
"grad_norm": 2.114203691482544,
|
263 |
+
"learning_rate": 1.740835464620631e-05,
|
264 |
+
"loss": 5.7936,
|
265 |
+
"step": 850
|
266 |
+
},
|
267 |
+
{
|
268 |
+
"epoch": 1.118925831202046,
|
269 |
+
"grad_norm": 1.8992066383361816,
|
270 |
+
"learning_rate": 1.7266268826371127e-05,
|
271 |
+
"loss": 5.7924,
|
272 |
+
"step": 875
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 1.1508951406649617,
|
276 |
+
"grad_norm": 2.8299577236175537,
|
277 |
+
"learning_rate": 1.7124183006535948e-05,
|
278 |
+
"loss": 5.7806,
|
279 |
+
"step": 900
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 1.1828644501278773,
|
283 |
+
"grad_norm": 1.956953525543213,
|
284 |
+
"learning_rate": 1.698209718670077e-05,
|
285 |
+
"loss": 5.7835,
|
286 |
+
"step": 925
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 1.2148337595907928,
|
290 |
+
"grad_norm": 2.658413887023926,
|
291 |
+
"learning_rate": 1.6840011366865587e-05,
|
292 |
+
"loss": 5.7619,
|
293 |
+
"step": 950
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 1.2468030690537084,
|
297 |
+
"grad_norm": 1.2760388851165771,
|
298 |
+
"learning_rate": 1.6697925547030408e-05,
|
299 |
+
"loss": 5.8038,
|
300 |
+
"step": 975
|
301 |
+
},
|
302 |
+
{
|
303 |
+
"epoch": 1.278772378516624,
|
304 |
+
"grad_norm": 1.7434897422790527,
|
305 |
+
"learning_rate": 1.6555839727195226e-05,
|
306 |
+
"loss": 5.779,
|
307 |
+
"step": 1000
|
308 |
+
},
|
309 |
+
{
|
310 |
+
"epoch": 1.3107416879795397,
|
311 |
+
"grad_norm": 1.3532071113586426,
|
312 |
+
"learning_rate": 1.6413753907360047e-05,
|
313 |
+
"loss": 5.7904,
|
314 |
+
"step": 1025
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 1.3427109974424551,
|
318 |
+
"grad_norm": 3.7385997772216797,
|
319 |
+
"learning_rate": 1.6271668087524864e-05,
|
320 |
+
"loss": 5.7696,
|
321 |
+
"step": 1050
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 1.3746803069053708,
|
325 |
+
"grad_norm": 0.9061102867126465,
|
326 |
+
"learning_rate": 1.6129582267689685e-05,
|
327 |
+
"loss": 5.7919,
|
328 |
+
"step": 1075
|
329 |
+
},
|
330 |
+
{
|
331 |
+
"epoch": 1.4066496163682864,
|
332 |
+
"grad_norm": 2.7104809284210205,
|
333 |
+
"learning_rate": 1.5987496447854503e-05,
|
334 |
+
"loss": 5.7785,
|
335 |
+
"step": 1100
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 1.438618925831202,
|
339 |
+
"grad_norm": 1.7147830724716187,
|
340 |
+
"learning_rate": 1.5845410628019324e-05,
|
341 |
+
"loss": 5.7862,
|
342 |
+
"step": 1125
|
343 |
+
},
|
344 |
+
{
|
345 |
+
"epoch": 1.4705882352941178,
|
346 |
+
"grad_norm": 2.525214672088623,
|
347 |
+
"learning_rate": 1.5703324808184145e-05,
|
348 |
+
"loss": 5.7703,
|
349 |
+
"step": 1150
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 1.5025575447570332,
|
353 |
+
"grad_norm": 1.7794997692108154,
|
354 |
+
"learning_rate": 1.5561238988348963e-05,
|
355 |
+
"loss": 5.773,
|
356 |
+
"step": 1175
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 1.5345268542199488,
|
360 |
+
"grad_norm": 4.901644229888916,
|
361 |
+
"learning_rate": 1.5419153168513784e-05,
|
362 |
+
"loss": 5.7627,
|
363 |
+
"step": 1200
|
364 |
+
},
|
365 |
+
{
|
366 |
+
"epoch": 1.5664961636828645,
|
367 |
+
"grad_norm": 3.360812187194824,
|
368 |
+
"learning_rate": 1.52770673486786e-05,
|
369 |
+
"loss": 5.7596,
|
370 |
+
"step": 1225
|
371 |
+
},
|
372 |
+
{
|
373 |
+
"epoch": 1.59846547314578,
|
374 |
+
"grad_norm": 1.2768888473510742,
|
375 |
+
"learning_rate": 1.5134981528843423e-05,
|
376 |
+
"loss": 5.7882,
|
377 |
+
"step": 1250
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 1.6304347826086958,
|
381 |
+
"grad_norm": 2.206226348876953,
|
382 |
+
"learning_rate": 1.4992895709008242e-05,
|
383 |
+
"loss": 5.7828,
|
384 |
+
"step": 1275
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 1.6624040920716112,
|
388 |
+
"grad_norm": 1.4602406024932861,
|
389 |
+
"learning_rate": 1.4850809889173061e-05,
|
390 |
+
"loss": 5.771,
|
391 |
+
"step": 1300
|
392 |
+
},
|
393 |
+
{
|
394 |
+
"epoch": 1.6943734015345269,
|
395 |
+
"grad_norm": 1.1597537994384766,
|
396 |
+
"learning_rate": 1.4708724069337881e-05,
|
397 |
+
"loss": 5.788,
|
398 |
+
"step": 1325
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 1.7263427109974425,
|
402 |
+
"grad_norm": 3.7494003772735596,
|
403 |
+
"learning_rate": 1.45666382495027e-05,
|
404 |
+
"loss": 5.7719,
|
405 |
+
"step": 1350
|
406 |
+
},
|
407 |
+
{
|
408 |
+
"epoch": 1.758312020460358,
|
409 |
+
"grad_norm": 1.6271498203277588,
|
410 |
+
"learning_rate": 1.442455242966752e-05,
|
411 |
+
"loss": 5.7846,
|
412 |
+
"step": 1375
|
413 |
+
},
|
414 |
+
{
|
415 |
+
"epoch": 1.7902813299232738,
|
416 |
+
"grad_norm": 2.0469117164611816,
|
417 |
+
"learning_rate": 1.4282466609832339e-05,
|
418 |
+
"loss": 5.7838,
|
419 |
+
"step": 1400
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 1.8222506393861893,
|
423 |
+
"grad_norm": 2.533921003341675,
|
424 |
+
"learning_rate": 1.4140380789997158e-05,
|
425 |
+
"loss": 5.7912,
|
426 |
+
"step": 1425
|
427 |
+
},
|
428 |
+
{
|
429 |
+
"epoch": 1.854219948849105,
|
430 |
+
"grad_norm": 3.291757583618164,
|
431 |
+
"learning_rate": 1.3998294970161978e-05,
|
432 |
+
"loss": 5.7686,
|
433 |
+
"step": 1450
|
434 |
+
},
|
435 |
+
{
|
436 |
+
"epoch": 1.8861892583120206,
|
437 |
+
"grad_norm": 3.0181350708007812,
|
438 |
+
"learning_rate": 1.3856209150326799e-05,
|
439 |
+
"loss": 5.7938,
|
440 |
+
"step": 1475
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 1.918158567774936,
|
444 |
+
"grad_norm": 2.553502321243286,
|
445 |
+
"learning_rate": 1.3714123330491618e-05,
|
446 |
+
"loss": 5.7847,
|
447 |
+
"step": 1500
|
448 |
+
},
|
449 |
+
{
|
450 |
+
"epoch": 1.9501278772378516,
|
451 |
+
"grad_norm": 1.8034719228744507,
|
452 |
+
"learning_rate": 1.3572037510656438e-05,
|
453 |
+
"loss": 5.7952,
|
454 |
+
"step": 1525
|
455 |
+
},
|
456 |
+
{
|
457 |
+
"epoch": 1.9820971867007673,
|
458 |
+
"grad_norm": 3.7138864994049072,
|
459 |
+
"learning_rate": 1.3429951690821257e-05,
|
460 |
+
"loss": 5.7528,
|
461 |
+
"step": 1550
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 2.0,
|
465 |
+
"eval_loss": 5.793323516845703,
|
466 |
+
"eval_runtime": 18.2796,
|
467 |
+
"eval_samples_per_second": 273.528,
|
468 |
+
"eval_steps_per_second": 4.322,
|
469 |
+
"eval_tr_ling_pearson_cosine": 0.037604255015168134,
|
470 |
+
"eval_tr_ling_pearson_dot": 0.0673696846368413,
|
471 |
+
"eval_tr_ling_pearson_euclidean": 0.03698411306484619,
|
472 |
+
"eval_tr_ling_pearson_manhattan": 0.034740275152181296,
|
473 |
+
"eval_tr_ling_pearson_max": 0.0673696846368413,
|
474 |
+
"eval_tr_ling_spearman_cosine": 0.04804112988506346,
|
475 |
+
"eval_tr_ling_spearman_dot": 0.06818119362900125,
|
476 |
+
"eval_tr_ling_spearman_euclidean": 0.03903062430281842,
|
477 |
+
"eval_tr_ling_spearman_manhattan": 0.03769766156967754,
|
478 |
+
"eval_tr_ling_spearman_max": 0.06818119362900125,
|
479 |
+
"step": 1564
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 2.0140664961636827,
|
483 |
+
"grad_norm": 2.8085248470306396,
|
484 |
+
"learning_rate": 1.3287865870986076e-05,
|
485 |
+
"loss": 5.65,
|
486 |
+
"step": 1575
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 2.0460358056265986,
|
490 |
+
"grad_norm": 3.3792033195495605,
|
491 |
+
"learning_rate": 1.3145780051150896e-05,
|
492 |
+
"loss": 5.7537,
|
493 |
+
"step": 1600
|
494 |
+
},
|
495 |
+
{
|
496 |
+
"epoch": 2.078005115089514,
|
497 |
+
"grad_norm": 3.44346022605896,
|
498 |
+
"learning_rate": 1.3003694231315715e-05,
|
499 |
+
"loss": 5.7098,
|
500 |
+
"step": 1625
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 2.10997442455243,
|
504 |
+
"grad_norm": 5.481964588165283,
|
505 |
+
"learning_rate": 1.2861608411480534e-05,
|
506 |
+
"loss": 5.7149,
|
507 |
+
"step": 1650
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 2.1419437340153453,
|
511 |
+
"grad_norm": 2.9816033840179443,
|
512 |
+
"learning_rate": 1.2719522591645354e-05,
|
513 |
+
"loss": 5.7585,
|
514 |
+
"step": 1675
|
515 |
+
},
|
516 |
+
{
|
517 |
+
"epoch": 2.1739130434782608,
|
518 |
+
"grad_norm": 3.2157652378082275,
|
519 |
+
"learning_rate": 1.2577436771810175e-05,
|
520 |
+
"loss": 5.7277,
|
521 |
+
"step": 1700
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 2.2058823529411766,
|
525 |
+
"grad_norm": 2.92006516456604,
|
526 |
+
"learning_rate": 1.2435350951974994e-05,
|
527 |
+
"loss": 5.7482,
|
528 |
+
"step": 1725
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 2.237851662404092,
|
532 |
+
"grad_norm": 3.7664051055908203,
|
533 |
+
"learning_rate": 1.2293265132139814e-05,
|
534 |
+
"loss": 5.7115,
|
535 |
+
"step": 1750
|
536 |
+
},
|
537 |
+
{
|
538 |
+
"epoch": 2.2698209718670075,
|
539 |
+
"grad_norm": 5.3445353507995605,
|
540 |
+
"learning_rate": 1.2151179312304633e-05,
|
541 |
+
"loss": 5.6895,
|
542 |
+
"step": 1775
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 2.3017902813299234,
|
546 |
+
"grad_norm": 4.100110054016113,
|
547 |
+
"learning_rate": 1.2009093492469452e-05,
|
548 |
+
"loss": 5.7389,
|
549 |
+
"step": 1800
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 2.333759590792839,
|
553 |
+
"grad_norm": 5.986413478851318,
|
554 |
+
"learning_rate": 1.1867007672634272e-05,
|
555 |
+
"loss": 5.7161,
|
556 |
+
"step": 1825
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 2.3657289002557547,
|
560 |
+
"grad_norm": 4.717130661010742,
|
561 |
+
"learning_rate": 1.1724921852799091e-05,
|
562 |
+
"loss": 5.7123,
|
563 |
+
"step": 1850
|
564 |
+
},
|
565 |
+
{
|
566 |
+
"epoch": 2.39769820971867,
|
567 |
+
"grad_norm": 2.833897352218628,
|
568 |
+
"learning_rate": 1.158283603296391e-05,
|
569 |
+
"loss": 5.7322,
|
570 |
+
"step": 1875
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 2.4296675191815855,
|
574 |
+
"grad_norm": 3.9461288452148438,
|
575 |
+
"learning_rate": 1.144075021312873e-05,
|
576 |
+
"loss": 5.7421,
|
577 |
+
"step": 1900
|
578 |
+
},
|
579 |
+
{
|
580 |
+
"epoch": 2.4616368286445014,
|
581 |
+
"grad_norm": 5.360823154449463,
|
582 |
+
"learning_rate": 1.1298664393293551e-05,
|
583 |
+
"loss": 5.7615,
|
584 |
+
"step": 1925
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 2.493606138107417,
|
588 |
+
"grad_norm": 3.290187120437622,
|
589 |
+
"learning_rate": 1.115657857345837e-05,
|
590 |
+
"loss": 5.7493,
|
591 |
+
"step": 1950
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 2.5255754475703327,
|
595 |
+
"grad_norm": 2.8723881244659424,
|
596 |
+
"learning_rate": 1.101449275362319e-05,
|
597 |
+
"loss": 5.7298,
|
598 |
+
"step": 1975
|
599 |
+
},
|
600 |
+
{
|
601 |
+
"epoch": 2.557544757033248,
|
602 |
+
"grad_norm": 12.763352394104004,
|
603 |
+
"learning_rate": 1.0872406933788009e-05,
|
604 |
+
"loss": 5.7529,
|
605 |
+
"step": 2000
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 2.5895140664961636,
|
609 |
+
"grad_norm": 3.423097610473633,
|
610 |
+
"learning_rate": 1.0730321113952828e-05,
|
611 |
+
"loss": 5.7318,
|
612 |
+
"step": 2025
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 2.6214833759590794,
|
616 |
+
"grad_norm": 3.546499252319336,
|
617 |
+
"learning_rate": 1.0588235294117648e-05,
|
618 |
+
"loss": 5.7036,
|
619 |
+
"step": 2050
|
620 |
+
},
|
621 |
+
{
|
622 |
+
"epoch": 2.653452685421995,
|
623 |
+
"grad_norm": 4.731326103210449,
|
624 |
+
"learning_rate": 1.0446149474282467e-05,
|
625 |
+
"loss": 5.7158,
|
626 |
+
"step": 2075
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 2.6854219948849103,
|
630 |
+
"grad_norm": 5.279483318328857,
|
631 |
+
"learning_rate": 1.0304063654447287e-05,
|
632 |
+
"loss": 5.7209,
|
633 |
+
"step": 2100
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 2.717391304347826,
|
637 |
+
"grad_norm": 5.814947605133057,
|
638 |
+
"learning_rate": 1.0161977834612106e-05,
|
639 |
+
"loss": 5.738,
|
640 |
+
"step": 2125
|
641 |
+
},
|
642 |
+
{
|
643 |
+
"epoch": 2.7493606138107416,
|
644 |
+
"grad_norm": 4.115816116333008,
|
645 |
+
"learning_rate": 1.0019892014776927e-05,
|
646 |
+
"loss": 5.7337,
|
647 |
+
"step": 2150
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 2.781329923273657,
|
651 |
+
"grad_norm": 4.176394462585449,
|
652 |
+
"learning_rate": 9.877806194941746e-06,
|
653 |
+
"loss": 5.713,
|
654 |
+
"step": 2175
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 2.813299232736573,
|
658 |
+
"grad_norm": 3.36919903755188,
|
659 |
+
"learning_rate": 9.735720375106566e-06,
|
660 |
+
"loss": 5.7257,
|
661 |
+
"step": 2200
|
662 |
+
},
|
663 |
+
{
|
664 |
+
"epoch": 2.8452685421994883,
|
665 |
+
"grad_norm": 4.4527482986450195,
|
666 |
+
"learning_rate": 9.593634555271385e-06,
|
667 |
+
"loss": 5.6958,
|
668 |
+
"step": 2225
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 2.877237851662404,
|
672 |
+
"grad_norm": 7.66256856918335,
|
673 |
+
"learning_rate": 9.451548735436205e-06,
|
674 |
+
"loss": 5.7053,
|
675 |
+
"step": 2250
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 2.9092071611253196,
|
679 |
+
"grad_norm": 11.90414810180664,
|
680 |
+
"learning_rate": 9.309462915601024e-06,
|
681 |
+
"loss": 5.7246,
|
682 |
+
"step": 2275
|
683 |
+
},
|
684 |
+
{
|
685 |
+
"epoch": 2.9411764705882355,
|
686 |
+
"grad_norm": 3.27648663520813,
|
687 |
+
"learning_rate": 9.167377095765843e-06,
|
688 |
+
"loss": 5.7291,
|
689 |
+
"step": 2300
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 2.973145780051151,
|
693 |
+
"grad_norm": 5.769582271575928,
|
694 |
+
"learning_rate": 9.025291275930663e-06,
|
695 |
+
"loss": 5.7139,
|
696 |
+
"step": 2325
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 3.0,
|
700 |
+
"eval_loss": 5.851009845733643,
|
701 |
+
"eval_runtime": 18.263,
|
702 |
+
"eval_samples_per_second": 273.777,
|
703 |
+
"eval_steps_per_second": 4.326,
|
704 |
+
"eval_tr_ling_pearson_cosine": 0.06129823646086187,
|
705 |
+
"eval_tr_ling_pearson_dot": 0.08667935948713909,
|
706 |
+
"eval_tr_ling_pearson_euclidean": 0.050963674624173616,
|
707 |
+
"eval_tr_ling_pearson_manhattan": 0.049471366228539336,
|
708 |
+
"eval_tr_ling_pearson_max": 0.08667935948713909,
|
709 |
+
"eval_tr_ling_spearman_cosine": 0.06262320788887717,
|
710 |
+
"eval_tr_ling_spearman_dot": 0.0836754651265069,
|
711 |
+
"eval_tr_ling_spearman_euclidean": 0.04874454654419082,
|
712 |
+
"eval_tr_ling_spearman_manhattan": 0.04780108900980343,
|
713 |
+
"eval_tr_ling_spearman_max": 0.0836754651265069,
|
714 |
+
"step": 2346
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 3.0051150895140664,
|
718 |
+
"grad_norm": 6.343133449554443,
|
719 |
+
"learning_rate": 8.883205456095482e-06,
|
720 |
+
"loss": 5.5715,
|
721 |
+
"step": 2350
|
722 |
+
},
|
723 |
+
{
|
724 |
+
"epoch": 3.0370843989769822,
|
725 |
+
"grad_norm": 7.939487457275391,
|
726 |
+
"learning_rate": 8.741119636260303e-06,
|
727 |
+
"loss": 5.6558,
|
728 |
+
"step": 2375
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 3.0690537084398977,
|
732 |
+
"grad_norm": 3.734879493713379,
|
733 |
+
"learning_rate": 8.599033816425122e-06,
|
734 |
+
"loss": 5.6441,
|
735 |
+
"step": 2400
|
736 |
+
},
|
737 |
+
{
|
738 |
+
"epoch": 3.101023017902813,
|
739 |
+
"grad_norm": 6.058401584625244,
|
740 |
+
"learning_rate": 8.456947996589942e-06,
|
741 |
+
"loss": 5.6569,
|
742 |
+
"step": 2425
|
743 |
+
},
|
744 |
+
{
|
745 |
+
"epoch": 3.132992327365729,
|
746 |
+
"grad_norm": 4.311662673950195,
|
747 |
+
"learning_rate": 8.314862176754761e-06,
|
748 |
+
"loss": 5.669,
|
749 |
+
"step": 2450
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 3.1649616368286444,
|
753 |
+
"grad_norm": 8.782428741455078,
|
754 |
+
"learning_rate": 8.17277635691958e-06,
|
755 |
+
"loss": 5.6361,
|
756 |
+
"step": 2475
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 3.1969309462915603,
|
760 |
+
"grad_norm": 7.427972793579102,
|
761 |
+
"learning_rate": 8.0306905370844e-06,
|
762 |
+
"loss": 5.6524,
|
763 |
+
"step": 2500
|
764 |
+
},
|
765 |
+
{
|
766 |
+
"epoch": 3.2289002557544757,
|
767 |
+
"grad_norm": 5.069025993347168,
|
768 |
+
"learning_rate": 7.88860471724922e-06,
|
769 |
+
"loss": 5.6773,
|
770 |
+
"step": 2525
|
771 |
+
},
|
772 |
+
{
|
773 |
+
"epoch": 3.260869565217391,
|
774 |
+
"grad_norm": 8.149388313293457,
|
775 |
+
"learning_rate": 7.746518897414039e-06,
|
776 |
+
"loss": 5.6552,
|
777 |
+
"step": 2550
|
778 |
+
},
|
779 |
+
{
|
780 |
+
"epoch": 3.292838874680307,
|
781 |
+
"grad_norm": 6.453441619873047,
|
782 |
+
"learning_rate": 7.604433077578858e-06,
|
783 |
+
"loss": 5.6807,
|
784 |
+
"step": 2575
|
785 |
+
},
|
786 |
+
{
|
787 |
+
"epoch": 3.3248081841432224,
|
788 |
+
"grad_norm": 6.5807719230651855,
|
789 |
+
"learning_rate": 7.4623472577436775e-06,
|
790 |
+
"loss": 5.6638,
|
791 |
+
"step": 2600
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 3.3567774936061383,
|
795 |
+
"grad_norm": 10.392335891723633,
|
796 |
+
"learning_rate": 7.320261437908497e-06,
|
797 |
+
"loss": 5.6582,
|
798 |
+
"step": 2625
|
799 |
+
},
|
800 |
+
{
|
801 |
+
"epoch": 3.3887468030690537,
|
802 |
+
"grad_norm": 9.251813888549805,
|
803 |
+
"learning_rate": 7.178175618073316e-06,
|
804 |
+
"loss": 5.658,
|
805 |
+
"step": 2650
|
806 |
+
},
|
807 |
+
{
|
808 |
+
"epoch": 3.420716112531969,
|
809 |
+
"grad_norm": 5.527411460876465,
|
810 |
+
"learning_rate": 7.036089798238136e-06,
|
811 |
+
"loss": 5.6626,
|
812 |
+
"step": 2675
|
813 |
+
},
|
814 |
+
{
|
815 |
+
"epoch": 3.452685421994885,
|
816 |
+
"grad_norm": 5.650461673736572,
|
817 |
+
"learning_rate": 6.894003978402956e-06,
|
818 |
+
"loss": 5.6802,
|
819 |
+
"step": 2700
|
820 |
+
},
|
821 |
+
{
|
822 |
+
"epoch": 3.4846547314578005,
|
823 |
+
"grad_norm": 7.156338691711426,
|
824 |
+
"learning_rate": 6.751918158567775e-06,
|
825 |
+
"loss": 5.6377,
|
826 |
+
"step": 2725
|
827 |
+
},
|
828 |
+
{
|
829 |
+
"epoch": 3.516624040920716,
|
830 |
+
"grad_norm": 6.843425750732422,
|
831 |
+
"learning_rate": 6.6098323387325946e-06,
|
832 |
+
"loss": 5.6752,
|
833 |
+
"step": 2750
|
834 |
+
},
|
835 |
+
{
|
836 |
+
"epoch": 3.5485933503836318,
|
837 |
+
"grad_norm": 14.204697608947754,
|
838 |
+
"learning_rate": 6.467746518897414e-06,
|
839 |
+
"loss": 5.6573,
|
840 |
+
"step": 2775
|
841 |
+
},
|
842 |
+
{
|
843 |
+
"epoch": 3.580562659846547,
|
844 |
+
"grad_norm": 3.9053664207458496,
|
845 |
+
"learning_rate": 6.325660699062234e-06,
|
846 |
+
"loss": 5.6963,
|
847 |
+
"step": 2800
|
848 |
+
},
|
849 |
+
{
|
850 |
+
"epoch": 3.612531969309463,
|
851 |
+
"grad_norm": 13.336016654968262,
|
852 |
+
"learning_rate": 6.1835748792270535e-06,
|
853 |
+
"loss": 5.7007,
|
854 |
+
"step": 2825
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 3.6445012787723785,
|
858 |
+
"grad_norm": 5.112432956695557,
|
859 |
+
"learning_rate": 6.041489059391873e-06,
|
860 |
+
"loss": 5.6746,
|
861 |
+
"step": 2850
|
862 |
+
},
|
863 |
+
{
|
864 |
+
"epoch": 3.6764705882352944,
|
865 |
+
"grad_norm": 6.077632427215576,
|
866 |
+
"learning_rate": 5.899403239556692e-06,
|
867 |
+
"loss": 5.6312,
|
868 |
+
"step": 2875
|
869 |
+
},
|
870 |
+
{
|
871 |
+
"epoch": 3.70843989769821,
|
872 |
+
"grad_norm": 10.304828643798828,
|
873 |
+
"learning_rate": 5.757317419721512e-06,
|
874 |
+
"loss": 5.5596,
|
875 |
+
"step": 2900
|
876 |
+
},
|
877 |
+
{
|
878 |
+
"epoch": 3.7404092071611252,
|
879 |
+
"grad_norm": 9.45308780670166,
|
880 |
+
"learning_rate": 5.615231599886332e-06,
|
881 |
+
"loss": 5.7003,
|
882 |
+
"step": 2925
|
883 |
+
},
|
884 |
+
{
|
885 |
+
"epoch": 3.772378516624041,
|
886 |
+
"grad_norm": 6.124211311340332,
|
887 |
+
"learning_rate": 5.473145780051151e-06,
|
888 |
+
"loss": 5.6739,
|
889 |
+
"step": 2950
|
890 |
+
},
|
891 |
+
{
|
892 |
+
"epoch": 3.8043478260869565,
|
893 |
+
"grad_norm": 8.547770500183105,
|
894 |
+
"learning_rate": 5.331059960215971e-06,
|
895 |
+
"loss": 5.655,
|
896 |
+
"step": 2975
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 3.836317135549872,
|
900 |
+
"grad_norm": 6.203834533691406,
|
901 |
+
"learning_rate": 5.18897414038079e-06,
|
902 |
+
"loss": 5.6787,
|
903 |
+
"step": 3000
|
904 |
+
},
|
905 |
+
{
|
906 |
+
"epoch": 3.868286445012788,
|
907 |
+
"grad_norm": 4.0565643310546875,
|
908 |
+
"learning_rate": 5.04688832054561e-06,
|
909 |
+
"loss": 5.643,
|
910 |
+
"step": 3025
|
911 |
+
},
|
912 |
+
{
|
913 |
+
"epoch": 3.9002557544757033,
|
914 |
+
"grad_norm": 9.590073585510254,
|
915 |
+
"learning_rate": 4.90480250071043e-06,
|
916 |
+
"loss": 5.6412,
|
917 |
+
"step": 3050
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 3.9322250639386187,
|
921 |
+
"grad_norm": 9.556587219238281,
|
922 |
+
"learning_rate": 4.762716680875249e-06,
|
923 |
+
"loss": 5.758,
|
924 |
+
"step": 3075
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 3.9641943734015346,
|
928 |
+
"grad_norm": 5.743387222290039,
|
929 |
+
"learning_rate": 4.620630861040068e-06,
|
930 |
+
"loss": 5.6769,
|
931 |
+
"step": 3100
|
932 |
+
},
|
933 |
+
{
|
934 |
+
"epoch": 3.99616368286445,
|
935 |
+
"grad_norm": 7.73360013961792,
|
936 |
+
"learning_rate": 4.478545041204888e-06,
|
937 |
+
"loss": 5.7206,
|
938 |
+
"step": 3125
|
939 |
+
},
|
940 |
+
{
|
941 |
+
"epoch": 4.0,
|
942 |
+
"eval_loss": 5.9124884605407715,
|
943 |
+
"eval_runtime": 18.3869,
|
944 |
+
"eval_samples_per_second": 271.933,
|
945 |
+
"eval_steps_per_second": 4.297,
|
946 |
+
"eval_tr_ling_pearson_cosine": 0.058743115070889876,
|
947 |
+
"eval_tr_ling_pearson_dot": 0.08477622619519222,
|
948 |
+
"eval_tr_ling_pearson_euclidean": 0.04709170917685587,
|
949 |
+
"eval_tr_ling_pearson_manhattan": 0.04582145815494953,
|
950 |
+
"eval_tr_ling_pearson_max": 0.08477622619519222,
|
951 |
+
"eval_tr_ling_spearman_cosine": 0.059526247945378225,
|
952 |
+
"eval_tr_ling_spearman_dot": 0.08243745050110735,
|
953 |
+
"eval_tr_ling_spearman_euclidean": 0.04407504959649961,
|
954 |
+
"eval_tr_ling_spearman_manhattan": 0.04331287037397966,
|
955 |
+
"eval_tr_ling_spearman_max": 0.08243745050110735,
|
956 |
+
"step": 3128
|
957 |
+
}
|
958 |
+
],
|
959 |
+
"logging_steps": 25,
|
960 |
+
"max_steps": 3910,
|
961 |
+
"num_input_tokens_seen": 0,
|
962 |
+
"num_train_epochs": 5,
|
963 |
+
"save_steps": 500,
|
964 |
+
"stateful_callbacks": {
|
965 |
+
"EarlyStoppingCallback": {
|
966 |
+
"args": {
|
967 |
+
"early_stopping_patience": 5,
|
968 |
+
"early_stopping_threshold": 0.01
|
969 |
+
},
|
970 |
+
"attributes": {
|
971 |
+
"early_stopping_patience_counter": 0
|
972 |
+
}
|
973 |
+
},
|
974 |
+
"TrainerControl": {
|
975 |
+
"args": {
|
976 |
+
"should_epoch_stop": false,
|
977 |
+
"should_evaluate": false,
|
978 |
+
"should_log": false,
|
979 |
+
"should_save": true,
|
980 |
+
"should_training_stop": false
|
981 |
+
},
|
982 |
+
"attributes": {}
|
983 |
+
}
|
984 |
+
},
|
985 |
+
"total_flos": 0.0,
|
986 |
+
"train_batch_size": 32,
|
987 |
+
"trial_name": null,
|
988 |
+
"trial_params": null
|
989 |
+
}
|
checkpoint-3128/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d44b0bec8869c08bdad0d597184d7a293a0e13eb770d6f4384456cbbe4fe5aa4
|
3 |
+
size 5368
|
checkpoint-3128/unigram.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:da145b5e7700ae40f16691ec32a0b1fdc1ee3298db22a31ea55f57a966c4a65d
|
3 |
+
size 14763260
|
config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2",
|
3 |
+
"architectures": [
|
4 |
+
"BertModel"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"classifier_dropout": null,
|
8 |
+
"gradient_checkpointing": false,
|
9 |
+
"hidden_act": "gelu",
|
10 |
+
"hidden_dropout_prob": 0.1,
|
11 |
+
"hidden_size": 384,
|
12 |
+
"initializer_range": 0.02,
|
13 |
+
"intermediate_size": 1536,
|
14 |
+
"layer_norm_eps": 1e-12,
|
15 |
+
"max_position_embeddings": 512,
|
16 |
+
"model_type": "bert",
|
17 |
+
"num_attention_heads": 12,
|
18 |
+
"num_hidden_layers": 12,
|
19 |
+
"pad_token_id": 0,
|
20 |
+
"position_embedding_type": "absolute",
|
21 |
+
"torch_dtype": "float32",
|
22 |
+
"transformers_version": "4.41.0",
|
23 |
+
"type_vocab_size": 2,
|
24 |
+
"use_cache": true,
|
25 |
+
"vocab_size": 250037
|
26 |
+
}
|
config_sentence_transformers.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"__version__": {
|
3 |
+
"sentence_transformers": "2.0.0",
|
4 |
+
"transformers": "4.7.0",
|
5 |
+
"pytorch": "1.9.0+cu102"
|
6 |
+
},
|
7 |
+
"prompts": {},
|
8 |
+
"default_prompt_name": null,
|
9 |
+
"similarity_fn_name": null
|
10 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4c62de6c84b1b2f1dd37271b7c328633d87e20fa33799d8acc6dd17e7276782d
|
3 |
+
size 470637416
|
modules.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
[
|
2 |
+
{
|
3 |
+
"idx": 0,
|
4 |
+
"name": "0",
|
5 |
+
"path": "",
|
6 |
+
"type": "sentence_transformers.models.Transformer"
|
7 |
+
},
|
8 |
+
{
|
9 |
+
"idx": 1,
|
10 |
+
"name": "1",
|
11 |
+
"path": "1_Pooling",
|
12 |
+
"type": "sentence_transformers.models.Pooling"
|
13 |
+
}
|
14 |
+
]
|
runs/May29_18-52-15_338a77628651/events.out.tfevents.1717008737.338a77628651.19835.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a1632c2aa6d9fe6916a4273d267eaecda68935aa9551a1745cdf8635a0fda945
|
3 |
+
size 11052
|
sentence_bert_config.json
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"max_seq_length": 128,
|
3 |
+
"do_lower_case": false
|
4 |
+
}
|
special_tokens_map.json
ADDED
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<s>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"cls_token": {
|
10 |
+
"content": "<s>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"eos_token": {
|
17 |
+
"content": "</s>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
},
|
23 |
+
"mask_token": {
|
24 |
+
"content": "<mask>",
|
25 |
+
"lstrip": true,
|
26 |
+
"normalized": false,
|
27 |
+
"rstrip": false,
|
28 |
+
"single_word": false
|
29 |
+
},
|
30 |
+
"pad_token": {
|
31 |
+
"content": "<pad>",
|
32 |
+
"lstrip": false,
|
33 |
+
"normalized": false,
|
34 |
+
"rstrip": false,
|
35 |
+
"single_word": false
|
36 |
+
},
|
37 |
+
"sep_token": {
|
38 |
+
"content": "</s>",
|
39 |
+
"lstrip": false,
|
40 |
+
"normalized": false,
|
41 |
+
"rstrip": false,
|
42 |
+
"single_word": false
|
43 |
+
},
|
44 |
+
"unk_token": {
|
45 |
+
"content": "<unk>",
|
46 |
+
"lstrip": false,
|
47 |
+
"normalized": false,
|
48 |
+
"rstrip": false,
|
49 |
+
"single_word": false
|
50 |
+
}
|
51 |
+
}
|
tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cad551d5600a84242d0973327029452a1e3672ba6313c2a3c3d69c4310e12719
|
3 |
+
size 17082987
|
tokenizer_config.json
ADDED
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "<s>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "<pad>",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "</s>",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"3": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"250001": {
|
36 |
+
"content": "<mask>",
|
37 |
+
"lstrip": true,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
}
|
43 |
+
},
|
44 |
+
"bos_token": "<s>",
|
45 |
+
"clean_up_tokenization_spaces": true,
|
46 |
+
"cls_token": "<s>",
|
47 |
+
"do_lower_case": true,
|
48 |
+
"eos_token": "</s>",
|
49 |
+
"mask_token": "<mask>",
|
50 |
+
"max_length": 128,
|
51 |
+
"model_max_length": 128,
|
52 |
+
"pad_to_multiple_of": null,
|
53 |
+
"pad_token": "<pad>",
|
54 |
+
"pad_token_type_id": 0,
|
55 |
+
"padding_side": "right",
|
56 |
+
"sep_token": "</s>",
|
57 |
+
"stride": 0,
|
58 |
+
"strip_accents": null,
|
59 |
+
"tokenize_chinese_chars": true,
|
60 |
+
"tokenizer_class": "BertTokenizer",
|
61 |
+
"truncation_side": "right",
|
62 |
+
"truncation_strategy": "longest_first",
|
63 |
+
"unk_token": "<unk>"
|
64 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:25f2c16c1654f713df68c72180bda7d273f3cdf401edb9ec64dfe4946a012590
|
3 |
+
size 5368
|
training_params.json
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"data_path": "ucsahin/TR-Extractive-QA-5K",
|
3 |
+
"model": "sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2",
|
4 |
+
"lr": 2e-05,
|
5 |
+
"epochs": 5,
|
6 |
+
"max_seq_length": 128,
|
7 |
+
"batch_size": 32,
|
8 |
+
"warmup_ratio": 0.1,
|
9 |
+
"gradient_accumulation": 1,
|
10 |
+
"optimizer": "adamw_torch",
|
11 |
+
"scheduler": "linear",
|
12 |
+
"weight_decay": 0.0,
|
13 |
+
"max_grad_norm": 1.0,
|
14 |
+
"seed": 42,
|
15 |
+
"train_split": "train",
|
16 |
+
"valid_split": null,
|
17 |
+
"logging_steps": -1,
|
18 |
+
"project_name": "turkish-embedding-model",
|
19 |
+
"auto_find_batch_size": false,
|
20 |
+
"mixed_precision": "fp16",
|
21 |
+
"save_total_limit": 1,
|
22 |
+
"push_to_hub": true,
|
23 |
+
"evaluation_strategy": "epoch",
|
24 |
+
"username": "acayir64",
|
25 |
+
"log": "tensorboard",
|
26 |
+
"early_stopping_patience": 5,
|
27 |
+
"early_stopping_threshold": 0.01,
|
28 |
+
"trainer": "qa",
|
29 |
+
"sentence1_column": "question",
|
30 |
+
"sentence2_column": "answer",
|
31 |
+
"sentence3_column": "sentence3",
|
32 |
+
"target_column": "target"
|
33 |
+
}
|
unigram.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:da145b5e7700ae40f16691ec32a0b1fdc1ee3298db22a31ea55f57a966c4a65d
|
3 |
+
size 14763260
|