Add new SentenceTransformer model

README.md

@@ -7,7 +7,7 @@ tags:
 - generated_from_trainer
 - dataset_size:90000
 - loss:MultipleNegativesRankingLoss
-base_model: thenlper/gte-small
+base_model: sentence-transformers/all-MiniLM-L6-v2
 widget:
 - source_sentence: what is the maximum i can contribute to a traditional ira
   sentences:
@@ -121,7 +121,7 @@ metrics:
 - cosine_mrr@10
 - cosine_map@100
 model-index:
-- name: SentenceTransformer based on thenlper/gte-small
+- name: SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2
   results:
   - task:
       type: information-retrieval
@@ -131,49 +131,49 @@ model-index:
       type: NanoMSMARCO
     metrics:
     - type: cosine_accuracy@1
-      value: 0.38
+      value: 0.3
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
-      value: 0.56
+      value: 0.62
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
-      value: 0.68
+      value: 0.62
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
       value: 0.72
       name: Cosine Accuracy@10
     - type: cosine_precision@1
-      value: 0.38
+      value: 0.3
       name: Cosine Precision@1
     - type: cosine_precision@3
-      value: 0.18666666666666668
+      value: 0.20666666666666667
       name: Cosine Precision@3
     - type: cosine_precision@5
-      value: 0.136
+      value: 0.124
       name: Cosine Precision@5
     - type: cosine_precision@10
       value: 0.07200000000000001
       name: Cosine Precision@10
     - type: cosine_recall@1
-      value: 0.38
+      value: 0.3
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.56
+      value: 0.62
       name: Cosine Recall@3
     - type: cosine_recall@5
-      value: 0.68
+      value: 0.62
       name: Cosine Recall@5
     - type: cosine_recall@10
       value: 0.72
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.5560017600077842
+      value: 0.5180879550984706
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.5023333333333333
+      value: 0.4529126984126984
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.5147876135794199
+      value: 0.4665743772173455
       name: Cosine Map@100
   - task:
       type: information-retrieval
@@ -183,49 +183,49 @@ model-index:
       type: NanoNQ
     metrics:
     - type: cosine_accuracy@1
-      value: 0.4
+      value: 0.36
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
-      value: 0.56
+      value: 0.5
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
-      value: 0.64
+      value: 0.54
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
-      value: 0.76
+      value: 0.6
       name: Cosine Accuracy@10
     - type: cosine_precision@1
-      value: 0.4
+      value: 0.36
       name: Cosine Precision@1
     - type: cosine_precision@3
-      value: 0.19333333333333333
+      value: 0.1733333333333333
       name: Cosine Precision@3
     - type: cosine_precision@5
-      value: 0.14
+      value: 0.11600000000000002
       name: Cosine Precision@5
     - type: cosine_precision@10
-      value: 0.08199999999999999
+      value: 0.066
       name: Cosine Precision@10
     - type: cosine_recall@1
-      value: 0.37
+      value: 0.34
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.54
+      value: 0.47
       name: Cosine Recall@3
     - type: cosine_recall@5
-      value: 0.63
+      value: 0.52
       name: Cosine Recall@5
     - type: cosine_recall@10
-      value: 0.74
+      value: 0.59
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.5529706231693197
+      value: 0.4736769259177555
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.505468253968254
+      value: 0.44483333333333336
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.4971957129558554
+      value: 0.4475757336608197
       name: Cosine Map@100
   - task:
       type: nano-beir
@@ -235,61 +235,61 @@ model-index:
       type: NanoBEIR_mean
     metrics:
     - type: cosine_accuracy@1
-      value: 0.39
+      value: 0.32999999999999996
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
       value: 0.56
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
-      value: 0.66
+      value: 0.5800000000000001
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
-      value: 0.74
+      value: 0.6599999999999999
       name: Cosine Accuracy@10
     - type: cosine_precision@1
-      value: 0.39
+      value: 0.32999999999999996
       name: Cosine Precision@1
     - type: cosine_precision@3
       value: 0.19
       name: Cosine Precision@3
     - type: cosine_precision@5
-      value: 0.138
+      value: 0.12000000000000001
       name: Cosine Precision@5
     - type: cosine_precision@10
-      value: 0.077
+      value: 0.069
       name: Cosine Precision@10
     - type: cosine_recall@1
-      value: 0.375
+      value: 0.32
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.55
+      value: 0.5449999999999999
       name: Cosine Recall@3
     - type: cosine_recall@5
-      value: 0.655
+      value: 0.5700000000000001
       name: Cosine Recall@5
     - type: cosine_recall@10
-      value: 0.73
+      value: 0.655
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.554486191588552
+      value: 0.49588244050811303
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.5039007936507937
+      value: 0.44887301587301587
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.5059916632676377
+      value: 0.4570750554390826
       name: Cosine Map@100
 ---
 
-# SentenceTransformer based on thenlper/gte-small
+# SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2
 
-This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [thenlper/gte-small](https://huggingface.co/thenlper/gte-small). 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.
+This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2). 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.
 
 ## Model Details
 
 ### Model Description
 - **Model Type:** Sentence Transformer
-- **Base model:** [thenlper/gte-small](https://huggingface.co/thenlper/gte-small) <!-- at revision 17e1f347d17fe144873b1201da91788898c639cd -->
+- **Base model:** [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) <!-- at revision c9745ed1d9f207416be6d2e6f8de32d1f16199bf -->
 - **Maximum Sequence Length:** 128 tokens
 - **Output Dimensionality:** 384 dimensions
 - **Similarity Function:** Cosine Similarity
@@ -342,9 +342,9 @@ print(embeddings.shape)
 # Get the similarity scores for the embeddings
 similarities = model.similarity(embeddings, embeddings)
 print(similarities)
-# tensor([[1.0000, 0.7010, 0.4682],
-#         [0.7010, 1.0000, 0.2656],
-#         [0.4682, 0.2656, 1.0001]])
+# tensor([[1.0000, 0.5217, 0.5263],
+#         [0.5217, 0.9999, 0.2880],
+#         [0.5263, 0.2880, 1.0000]])
 ```
 
 <!--
@@ -380,23 +380,23 @@ You can finetune this model on your own dataset.
 * Datasets: `NanoMSMARCO` and `NanoNQ`
 * Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
 
-| Metric              | NanoMSMARCO | NanoNQ    |
-|:--------------------|:------------|:----------|
-| cosine_accuracy@1   | 0.38        | 0.4       |
-| cosine_accuracy@3   | 0.56        | 0.56      |
-| cosine_accuracy@5   | 0.68        | 0.64      |
-| cosine_accuracy@10  | 0.72        | 0.76      |
-| cosine_precision@1  | 0.38        | 0.4       |
-| cosine_precision@3  | 0.1867      | 0.1933    |
-| cosine_precision@5  | 0.136       | 0.14      |
-| cosine_precision@10 | 0.072       | 0.082     |
-| cosine_recall@1     | 0.38        | 0.37      |
-| cosine_recall@3     | 0.56        | 0.54      |
-| cosine_recall@5     | 0.68        | 0.63      |
-| cosine_recall@10    | 0.72        | 0.74      |
-| **cosine_ndcg@10**  | **0.556**   | **0.553** |
-| cosine_mrr@10       | 0.5023      | 0.5055    |
-| cosine_map@100      | 0.5148      | 0.4972    |
+| Metric              | NanoMSMARCO | NanoNQ     |
+|:--------------------|:------------|:-----------|
+| cosine_accuracy@1   | 0.3         | 0.36       |
+| cosine_accuracy@3   | 0.62        | 0.5        |
+| cosine_accuracy@5   | 0.62        | 0.54       |
+| cosine_accuracy@10  | 0.72        | 0.6        |
+| cosine_precision@1  | 0.3         | 0.36       |
+| cosine_precision@3  | 0.2067      | 0.1733     |
+| cosine_precision@5  | 0.124       | 0.116      |
+| cosine_precision@10 | 0.072       | 0.066      |
+| cosine_recall@1     | 0.3         | 0.34       |
+| cosine_recall@3     | 0.62        | 0.47       |
+| cosine_recall@5     | 0.62        | 0.52       |
+| cosine_recall@10    | 0.72        | 0.59       |
+| **cosine_ndcg@10**  | **0.5181**  | **0.4737** |
+| cosine_mrr@10       | 0.4529      | 0.4448     |
+| cosine_map@100      | 0.4666      | 0.4476     |
 
 #### Nano BEIR
 
@@ -414,21 +414,21 @@ You can finetune this model on your own dataset.
 
 | Metric              | Value      |
 |:--------------------|:-----------|
-| cosine_accuracy@1   | 0.39       |
+| cosine_accuracy@1   | 0.33       |
 | cosine_accuracy@3   | 0.56       |
-| cosine_accuracy@5   | 0.66       |
-| cosine_accuracy@10  | 0.74       |
-| cosine_precision@1  | 0.39       |
+| cosine_accuracy@5   | 0.58       |
+| cosine_accuracy@10  | 0.66       |
+| cosine_precision@1  | 0.33       |
 | cosine_precision@3  | 0.19       |
-| cosine_precision@5  | 0.138      |
-| cosine_precision@10 | 0.077      |
-| cosine_recall@1     | 0.375      |
-| cosine_recall@3     | 0.55       |
-| cosine_recall@5     | 0.655      |
-| cosine_recall@10    | 0.73       |
-| **cosine_ndcg@10**  | **0.5545** |
-| cosine_mrr@10       | 0.5039     |
-| cosine_map@100      | 0.506      |
+| cosine_precision@5  | 0.12       |
+| cosine_precision@10 | 0.069      |
+| cosine_recall@1     | 0.32       |
+| cosine_recall@3     | 0.545      |
+| cosine_recall@5     | 0.57       |
+| cosine_recall@10    | 0.655      |
+| **cosine_ndcg@10**  | **0.4959** |
+| cosine_mrr@10       | 0.4489     |
+| cosine_map@100      | 0.4571     |
 
 <!--
 ## Bias, Risks and Limitations
@@ -502,9 +502,9 @@ You can finetune this model on your own dataset.
 - `eval_strategy`: steps
 - `per_device_train_batch_size`: 128
 - `per_device_eval_batch_size`: 128
-- `learning_rate`: 1e-05
-- `weight_decay`: 0.001
-- `max_steps`: 3000
+- `learning_rate`: 0.0001
+- `weight_decay`: 0.005
+- `max_steps`: 562
 - `warmup_ratio`: 0.1
 - `fp16`: True
 - `dataloader_drop_last`: True
@@ -531,14 +531,14 @@ You can finetune this model on your own dataset.
 - `gradient_accumulation_steps`: 1
 - `eval_accumulation_steps`: None
 - `torch_empty_cache_steps`: None
-- `learning_rate`: 1e-05
-- `weight_decay`: 0.001
+- `learning_rate`: 0.0001
+- `weight_decay`: 0.005
 - `adam_beta1`: 0.9
 - `adam_beta2`: 0.999
 - `adam_epsilon`: 1e-08
 - `max_grad_norm`: 1.0
 - `num_train_epochs`: 3.0
-- `max_steps`: 3000
+- `max_steps`: 562
 - `lr_scheduler_type`: linear
 - `lr_scheduler_kwargs`: {}
 - `warmup_ratio`: 0.1
@@ -645,19 +645,9 @@ You can finetune this model on your own dataset.
 ### Training Logs
 | Epoch  | Step | Training Loss | Validation Loss | NanoMSMARCO_cosine_ndcg@10 | NanoNQ_cosine_ndcg@10 | NanoBEIR_mean_cosine_ndcg@10 |
 |:------:|:----:|:-------------:|:---------------:|:--------------------------:|:---------------------:|:----------------------------:|
-| 0      | 0    | -             | 4.1735          | 0.6259                     | 0.6583                | 0.6421                       |
-| 0.3556 | 250  | 3.3399        | 1.1023          | 0.5061                     | 0.5036                | 0.5048                       |
-| 0.7112 | 500  | 1.2152        | 0.9648          | 0.5609                     | 0.5241                | 0.5425                       |
-| 1.0669 | 750  | 1.1157        | 0.9385          | 0.5568                     | 0.5444                | 0.5506                       |
-| 1.4225 | 1000 | 1.0721        | 0.9241          | 0.5523                     | 0.5520                | 0.5521                       |
-| 1.7781 | 1250 | 1.0561        | 0.9159          | 0.5677                     | 0.5454                | 0.5565                       |
-| 2.1337 | 1500 | 1.0411        | 0.9116          | 0.5505                     | 0.5527                | 0.5516                       |
-| 2.4893 | 1750 | 1.0248        | 0.9073          | 0.5433                     | 0.5631                | 0.5532                       |
-| 2.8450 | 2000 | 1.022         | 0.9042          | 0.5424                     | 0.5648                | 0.5536                       |
-| 3.2006 | 2250 | 1.0079        | 0.9015          | 0.5557                     | 0.5626                | 0.5592                       |
-| 3.5562 | 2500 | 1.0092        | 0.8998          | 0.5560                     | 0.5590                | 0.5575                       |
-| 3.9118 | 2750 | 0.9982        | 0.8986          | 0.5560                     | 0.5535                | 0.5548                       |
-| 4.2674 | 3000 | 0.9947        | 0.8986          | 0.5560                     | 0.5530                | 0.5545                       |
+| 0      | 0    | -             | 1.2542          | 0.5540                     | 0.5931                | 0.5735                       |
+| 0.3556 | 250  | 1.138         | 0.9359          | 0.5039                     | 0.4586                | 0.4813                       |
+| 0.7112 | 500  | 1.0665        | 0.9102          | 0.5181                     | 0.4737                | 0.4959                       |
 
 
 ### Framework Versions

config_sentence_transformers.json

@@ -1,10 +1,10 @@
 {
-  "model_type": "SentenceTransformer",
   "__version__": {
     "sentence_transformers": "5.2.0",
     "transformers": "4.57.3",
     "pytorch": "2.9.1+cu128"
   },
+  "model_type": "SentenceTransformer",
   "prompts": {
     "query": "",
     "document": ""