Add new SentenceTransformer model

README.md

@@ -7,7 +7,7 @@ tags:
 - generated_from_trainer
 - dataset_size:111470
 - loss:MultipleNegativesRankingLoss
-base_model: thenlper/gte-small
+base_model: sentence-transformers/all-MiniLM-L6-v2
 widget:
 - source_sentence: why are some rocks radioactive
   sentences:
@@ -106,7 +106,7 @@ 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
@@ -119,46 +119,46 @@ model-index:
       value: 0.32
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
-      value: 0.56
+      value: 0.5
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
-      value: 0.68
+      value: 0.56
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
-      value: 0.74
+      value: 0.72
       name: Cosine Accuracy@10
     - type: cosine_precision@1
       value: 0.32
       name: Cosine Precision@1
     - type: cosine_precision@3
-      value: 0.18666666666666668
+      value: 0.16666666666666669
       name: Cosine Precision@3
     - type: cosine_precision@5
-      value: 0.136
+      value: 0.11200000000000002
       name: Cosine Precision@5
     - type: cosine_precision@10
-      value: 0.07400000000000001
+      value: 0.07200000000000001
       name: Cosine Precision@10
     - type: cosine_recall@1
       value: 0.32
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.56
+      value: 0.5
       name: Cosine Recall@3
     - type: cosine_recall@5
-      value: 0.68
+      value: 0.56
       name: Cosine Recall@5
     - type: cosine_recall@10
-      value: 0.74
+      value: 0.72
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.536962042440356
+      value: 0.5076844979819899
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.47152380952380946
+      value: 0.4414682539682539
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.4836600263506265
+      value: 0.4541878759718346
       name: Cosine Map@100
   - task:
       type: information-retrieval
@@ -168,49 +168,49 @@ model-index:
       type: NanoNQ
     metrics:
     - type: cosine_accuracy@1
-      value: 0.42
+      value: 0.32
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
-      value: 0.56
+      value: 0.48
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
-      value: 0.6
+      value: 0.52
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
-      value: 0.74
+      value: 0.54
       name: Cosine Accuracy@10
     - type: cosine_precision@1
-      value: 0.42
+      value: 0.32
       name: Cosine Precision@1
     - type: cosine_precision@3
-      value: 0.2
+      value: 0.16666666666666663
       name: Cosine Precision@3
     - type: cosine_precision@5
-      value: 0.132
+      value: 0.10800000000000001
       name: Cosine Precision@5
     - type: cosine_precision@10
-      value: 0.08
+      value: 0.05800000000000001
       name: Cosine Precision@10
     - type: cosine_recall@1
-      value: 0.39
+      value: 0.29
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.54
+      value: 0.45
       name: Cosine Recall@3
     - type: cosine_recall@5
-      value: 0.59
+      value: 0.49
       name: Cosine Recall@5
     - type: cosine_recall@10
-      value: 0.72
+      value: 0.52
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.5505596490405809
+      value: 0.4230776979752646
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.5070793650793651
+      value: 0.40852380952380957
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.50321316661942
+      value: 0.4024677771121777
       name: Cosine Map@100
   - task:
       type: nano-beir
@@ -220,61 +220,61 @@ model-index:
       type: NanoBEIR_mean
     metrics:
     - type: cosine_accuracy@1
-      value: 0.37
+      value: 0.32
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
-      value: 0.56
+      value: 0.49
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
-      value: 0.64
+      value: 0.54
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
-      value: 0.74
+      value: 0.63
       name: Cosine Accuracy@10
     - type: cosine_precision@1
-      value: 0.37
+      value: 0.32
       name: Cosine Precision@1
     - type: cosine_precision@3
-      value: 0.19333333333333336
+      value: 0.16666666666666666
       name: Cosine Precision@3
     - type: cosine_precision@5
-      value: 0.134
+      value: 0.11000000000000001
       name: Cosine Precision@5
     - type: cosine_precision@10
-      value: 0.07700000000000001
+      value: 0.065
       name: Cosine Precision@10
     - type: cosine_recall@1
-      value: 0.355
+      value: 0.305
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.55
+      value: 0.475
       name: Cosine Recall@3
     - type: cosine_recall@5
-      value: 0.635
+      value: 0.525
       name: Cosine Recall@5
     - type: cosine_recall@10
-      value: 0.73
+      value: 0.62
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.5437608457404683
+      value: 0.46538109797862726
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.4893015873015873
+      value: 0.42499603174603173
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.49343659648502325
+      value: 0.42832782654200613
       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
@@ -327,9 +327,9 @@ print(embeddings.shape)
 # Get the similarity scores for the embeddings
 similarities = model.similarity(embeddings, embeddings)
 print(similarities)
-# tensor([[1.0000, 1.0000, 0.9823],
-#         [1.0000, 1.0000, 0.9823],
-#         [0.9823, 0.9823, 1.0000]])
+# tensor([[1.0000, 1.0000, 0.9805],
+#         [1.0000, 1.0000, 0.9805],
+#         [0.9805, 0.9805, 1.0000]])
 ```
 
 <!--
@@ -367,21 +367,21 @@ You can finetune this model on your own dataset.
 
 | Metric              | NanoMSMARCO | NanoNQ     |
 |:--------------------|:------------|:-----------|
-| cosine_accuracy@1   | 0.32        | 0.42       |
-| cosine_accuracy@3   | 0.56        | 0.56       |
-| cosine_accuracy@5   | 0.68        | 0.6        |
-| cosine_accuracy@10  | 0.74        | 0.74       |
-| cosine_precision@1  | 0.32        | 0.42       |
-| cosine_precision@3  | 0.1867      | 0.2        |
-| cosine_precision@5  | 0.136       | 0.132      |
-| cosine_precision@10 | 0.074       | 0.08       |
-| cosine_recall@1     | 0.32        | 0.39       |
-| cosine_recall@3     | 0.56        | 0.54       |
-| cosine_recall@5     | 0.68        | 0.59       |
-| cosine_recall@10    | 0.74        | 0.72       |
-| **cosine_ndcg@10**  | **0.537**   | **0.5506** |
-| cosine_mrr@10       | 0.4715      | 0.5071     |
-| cosine_map@100      | 0.4837      | 0.5032     |
+| cosine_accuracy@1   | 0.32        | 0.32       |
+| cosine_accuracy@3   | 0.5         | 0.48       |
+| cosine_accuracy@5   | 0.56        | 0.52       |
+| cosine_accuracy@10  | 0.72        | 0.54       |
+| cosine_precision@1  | 0.32        | 0.32       |
+| cosine_precision@3  | 0.1667      | 0.1667     |
+| cosine_precision@5  | 0.112       | 0.108      |
+| cosine_precision@10 | 0.072       | 0.058      |
+| cosine_recall@1     | 0.32        | 0.29       |
+| cosine_recall@3     | 0.5         | 0.45       |
+| cosine_recall@5     | 0.56        | 0.49       |
+| cosine_recall@10    | 0.72        | 0.52       |
+| **cosine_ndcg@10**  | **0.5077**  | **0.4231** |
+| cosine_mrr@10       | 0.4415      | 0.4085     |
+| cosine_map@100      | 0.4542      | 0.4025     |
 
 #### Nano BEIR
 
@@ -399,21 +399,21 @@ You can finetune this model on your own dataset.
 
 | Metric              | Value      |
 |:--------------------|:-----------|
-| cosine_accuracy@1   | 0.37       |
-| cosine_accuracy@3   | 0.56       |
-| cosine_accuracy@5   | 0.64       |
-| cosine_accuracy@10  | 0.74       |
-| cosine_precision@1  | 0.37       |
-| cosine_precision@3  | 0.1933     |
-| cosine_precision@5  | 0.134      |
-| cosine_precision@10 | 0.077      |
-| cosine_recall@1     | 0.355      |
-| cosine_recall@3     | 0.55       |
-| cosine_recall@5     | 0.635      |
-| cosine_recall@10    | 0.73       |
-| **cosine_ndcg@10**  | **0.5438** |
-| cosine_mrr@10       | 0.4893     |
-| cosine_map@100      | 0.4934     |
+| cosine_accuracy@1   | 0.32       |
+| cosine_accuracy@3   | 0.49       |
+| cosine_accuracy@5   | 0.54       |
+| cosine_accuracy@10  | 0.63       |
+| cosine_precision@1  | 0.32       |
+| cosine_precision@3  | 0.1667     |
+| cosine_precision@5  | 0.11       |
+| cosine_precision@10 | 0.065      |
+| cosine_recall@1     | 0.305      |
+| cosine_recall@3     | 0.475      |
+| cosine_recall@5     | 0.525      |
+| cosine_recall@10    | 0.62       |
+| **cosine_ndcg@10**  | **0.4654** |
+| cosine_mrr@10       | 0.425      |
+| cosine_map@100      | 0.4283     |
 
 <!--
 ## Bias, Risks and Limitations
@@ -487,9 +487,9 @@ 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`: 8e-05
+- `learning_rate`: 0.0001
 - `weight_decay`: 0.005
-- `max_steps`: 500
+- `max_steps`: 2250
 - `warmup_ratio`: 0.1
 - `fp16`: True
 - `dataloader_drop_last`: True
@@ -516,14 +516,14 @@ You can finetune this model on your own dataset.
 - `gradient_accumulation_steps`: 1
 - `eval_accumulation_steps`: None
 - `torch_empty_cache_steps`: None
-- `learning_rate`: 8e-05
+- `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`: 500
+- `max_steps`: 2250
 - `lr_scheduler_type`: linear
 - `lr_scheduler_kwargs`: {}
 - `warmup_ratio`: 0.1
@@ -628,13 +628,19 @@ You can finetune this model on your own dataset.
 </details>
 
 ### Training Logs
-| Epoch      | Step    | Training Loss | Validation Loss | NanoMSMARCO_cosine_ndcg@10 | NanoNQ_cosine_ndcg@10 | NanoBEIR_mean_cosine_ndcg@10 |
-|:----------:|:-------:|:-------------:|:---------------:|:--------------------------:|:---------------------:|:----------------------------:|
-| 0          | 0       | -             | 4.0678          | 0.6259                     | 0.6583                | 0.6421                       |
-| 0.2874     | 250     | 1.4121        | 0.8381          | 0.5267                     | 0.5321                | 0.5294                       |
-| **0.5747** | **500** | **0.9465**    | **0.8126**      | **0.537**                  | **0.5506**            | **0.5438**                   |
+| Epoch  | Step | Training Loss | Validation Loss | NanoMSMARCO_cosine_ndcg@10 | NanoNQ_cosine_ndcg@10 | NanoBEIR_mean_cosine_ndcg@10 |
+|:------:|:----:|:-------------:|:---------------:|:--------------------------:|:---------------------:|:----------------------------:|
+| 0      | 0    | -             | 1.1445          | 0.5540                     | 0.5931                | 0.5735                       |
+| 0.2874 | 250  | 1.0772        | 0.8647          | 0.4705                     | 0.5045                | 0.4875                       |
+| 0.5747 | 500  | 0.9853        | 0.8353          | 0.4884                     | 0.4658                | 0.4771                       |
+| 0.8621 | 750  | 0.9571        | 0.8153          | 0.5181                     | 0.4371                | 0.4776                       |
+| 1.1494 | 1000 | 0.8923        | 0.8060          | 0.4848                     | 0.4274                | 0.4561                       |
+| 1.4368 | 1250 | 0.8458        | 0.8020          | 0.5415                     | 0.4466                | 0.4941                       |
+| 1.7241 | 1500 | 0.8394        | 0.7929          | 0.4928                     | 0.4569                | 0.4748                       |
+| 2.0115 | 1750 | 0.8308        | 0.7917          | 0.5332                     | 0.4285                | 0.4808                       |
+| 2.2989 | 2000 | 0.7709        | 0.7921          | 0.4952                     | 0.4088                | 0.4520                       |
+| 2.5862 | 2250 | 0.7685        | 0.7914          | 0.5077                     | 0.4231                | 0.4654                       |
 
-* The bold row denotes the saved checkpoint.
 
 ### Framework Versions
 - Python: 3.10.18

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": ""