Add new SentenceTransformer model

README.md

@@ -7,7 +7,7 @@ tags:
 - generated_from_trainer
 - dataset_size:111470
 - loss:MultipleNegativesRankingLoss
-base_model: sentence-transformers/all-MiniLM-L6-v2
+base_model: sentence-transformers/all-MiniLM-L12-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 sentence-transformers/all-MiniLM-L6-v2
+- name: SentenceTransformer based on sentence-transformers/all-MiniLM-L12-v2
   results:
   - task:
       type: information-retrieval
@@ -116,49 +116,49 @@ model-index:
       type: NanoMSMARCO
     metrics:
     - type: cosine_accuracy@1
-      value: 0.34
+      value: 0.36
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
-      value: 0.5
+      value: 0.54
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
-      value: 0.6
+      value: 0.62
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
-      value: 0.7
+      value: 0.74
       name: Cosine Accuracy@10
     - type: cosine_precision@1
-      value: 0.34
+      value: 0.36
       name: Cosine Precision@1
     - type: cosine_precision@3
-      value: 0.16666666666666669
+      value: 0.18
       name: Cosine Precision@3
     - type: cosine_precision@5
-      value: 0.12000000000000002
+      value: 0.124
       name: Cosine Precision@5
     - type: cosine_precision@10
-      value: 0.07
+      value: 0.07400000000000001
       name: Cosine Precision@10
     - type: cosine_recall@1
-      value: 0.34
+      value: 0.36
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.5
+      value: 0.54
       name: Cosine Recall@3
     - type: cosine_recall@5
-      value: 0.6
+      value: 0.62
       name: Cosine Recall@5
     - type: cosine_recall@10
-      value: 0.7
+      value: 0.74
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.5086414353894502
+      value: 0.5388909110035328
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.4489126984126983
+      value: 0.4758571428571428
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.4629036994571799
+      value: 0.4857996035129028
       name: Cosine Map@100
   - task:
       type: information-retrieval
@@ -168,49 +168,49 @@ model-index:
       type: NanoNQ
     metrics:
     - type: cosine_accuracy@1
-      value: 0.4
+      value: 0.38
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
-      value: 0.54
+      value: 0.56
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
-      value: 0.56
+      value: 0.6
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
-      value: 0.6
+      value: 0.66
       name: Cosine Accuracy@10
     - type: cosine_precision@1
-      value: 0.4
+      value: 0.38
       name: Cosine Precision@1
     - type: cosine_precision@3
-      value: 0.19333333333333333
+      value: 0.2
       name: Cosine Precision@3
     - type: cosine_precision@5
-      value: 0.12400000000000003
+      value: 0.128
       name: Cosine Precision@5
     - type: cosine_precision@10
-      value: 0.066
+      value: 0.07
       name: Cosine Precision@10
     - type: cosine_recall@1
-      value: 0.36
+      value: 0.35
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.52
+      value: 0.54
       name: Cosine Recall@3
     - type: cosine_recall@5
-      value: 0.55
+      value: 0.58
       name: Cosine Recall@5
     - type: cosine_recall@10
-      value: 0.59
+      value: 0.64
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.49563614338260625
+      value: 0.5086947283606115
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.4761111111111111
+      value: 0.4757142857142857
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.47880687529822674
+      value: 0.4762724913936289
       name: Cosine Map@100
   - task:
       type: nano-beir
@@ -223,58 +223,58 @@ model-index:
       value: 0.37
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
-      value: 0.52
+      value: 0.55
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
-      value: 0.5800000000000001
+      value: 0.61
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
-      value: 0.6499999999999999
+      value: 0.7
       name: Cosine Accuracy@10
     - type: cosine_precision@1
       value: 0.37
       name: Cosine Precision@1
     - type: cosine_precision@3
-      value: 0.18
+      value: 0.19
       name: Cosine Precision@3
     - type: cosine_precision@5
-      value: 0.12200000000000003
+      value: 0.126
       name: Cosine Precision@5
     - type: cosine_precision@10
-      value: 0.068
+      value: 0.07200000000000001
       name: Cosine Precision@10
     - type: cosine_recall@1
-      value: 0.35
+      value: 0.355
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.51
+      value: 0.54
       name: Cosine Recall@3
     - type: cosine_recall@5
-      value: 0.575
+      value: 0.6
       name: Cosine Recall@5
     - type: cosine_recall@10
-      value: 0.645
+      value: 0.69
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.5021387893860282
+      value: 0.5237928196820721
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.46251190476190474
+      value: 0.47578571428571426
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.4708552873777033
+      value: 0.4810360474532659
       name: Cosine Map@100
 ---
 
-# SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2
+# SentenceTransformer based on sentence-transformers/all-MiniLM-L12-v2
 
-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.
+This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/all-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L12-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:** [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) <!-- at revision c9745ed1d9f207416be6d2e6f8de32d1f16199bf -->
+- **Base model:** [sentence-transformers/all-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L12-v2) <!-- at revision 936af83a2ecce5fe87a09109ff5cbcefe073173a -->
 - **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, 0.9770],
-#         [1.0000, 1.0000, 0.9770],
-#         [0.9770, 0.9770, 1.0000]])
+# tensor([[1.0001, 1.0001, 0.9818],
+#         [1.0001, 1.0001, 0.9818],
+#         [0.9818, 0.9818, 0.9999]])
 ```
 
 <!--
@@ -367,21 +367,21 @@ You can finetune this model on your own dataset.
 
 | Metric              | NanoMSMARCO | NanoNQ     |
 |:--------------------|:------------|:-----------|
-| cosine_accuracy@1   | 0.34        | 0.4        |
-| cosine_accuracy@3   | 0.5         | 0.54       |
-| cosine_accuracy@5   | 0.6         | 0.56       |
-| cosine_accuracy@10  | 0.7         | 0.6        |
-| cosine_precision@1  | 0.34        | 0.4        |
-| cosine_precision@3  | 0.1667      | 0.1933     |
-| cosine_precision@5  | 0.12        | 0.124      |
-| cosine_precision@10 | 0.07        | 0.066      |
-| cosine_recall@1     | 0.34        | 0.36       |
-| cosine_recall@3     | 0.5         | 0.52       |
-| cosine_recall@5     | 0.6         | 0.55       |
-| cosine_recall@10    | 0.7         | 0.59       |
-| **cosine_ndcg@10**  | **0.5086**  | **0.4956** |
-| cosine_mrr@10       | 0.4489      | 0.4761     |
-| cosine_map@100      | 0.4629      | 0.4788     |
+| cosine_accuracy@1   | 0.36        | 0.38       |
+| cosine_accuracy@3   | 0.54        | 0.56       |
+| cosine_accuracy@5   | 0.62        | 0.6        |
+| cosine_accuracy@10  | 0.74        | 0.66       |
+| cosine_precision@1  | 0.36        | 0.38       |
+| cosine_precision@3  | 0.18        | 0.2        |
+| cosine_precision@5  | 0.124       | 0.128      |
+| cosine_precision@10 | 0.074       | 0.07       |
+| cosine_recall@1     | 0.36        | 0.35       |
+| cosine_recall@3     | 0.54        | 0.54       |
+| cosine_recall@5     | 0.62        | 0.58       |
+| cosine_recall@10    | 0.74        | 0.64       |
+| **cosine_ndcg@10**  | **0.5389**  | **0.5087** |
+| cosine_mrr@10       | 0.4759      | 0.4757     |
+| cosine_map@100      | 0.4858      | 0.4763     |
 
 #### Nano BEIR
 
@@ -400,20 +400,20 @@ You can finetune this model on your own dataset.
 | Metric              | Value      |
 |:--------------------|:-----------|
 | cosine_accuracy@1   | 0.37       |
-| cosine_accuracy@3   | 0.52       |
-| cosine_accuracy@5   | 0.58       |
-| cosine_accuracy@10  | 0.65       |
+| cosine_accuracy@3   | 0.55       |
+| cosine_accuracy@5   | 0.61       |
+| cosine_accuracy@10  | 0.7        |
 | cosine_precision@1  | 0.37       |
-| cosine_precision@3  | 0.18       |
-| cosine_precision@5  | 0.122      |
-| cosine_precision@10 | 0.068      |
-| cosine_recall@1     | 0.35       |
-| cosine_recall@3     | 0.51       |
-| cosine_recall@5     | 0.575      |
-| cosine_recall@10    | 0.645      |
-| **cosine_ndcg@10**  | **0.5021** |
-| cosine_mrr@10       | 0.4625     |
-| cosine_map@100      | 0.4709     |
+| cosine_precision@3  | 0.19       |
+| cosine_precision@5  | 0.126      |
+| cosine_precision@10 | 0.072      |
+| cosine_recall@1     | 0.355      |
+| cosine_recall@3     | 0.54       |
+| cosine_recall@5     | 0.6        |
+| cosine_recall@10    | 0.69       |
+| **cosine_ndcg@10**  | **0.5238** |
+| cosine_mrr@10       | 0.4758     |
+| cosine_map@100      | 0.481      |
 
 <!--
 ## 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`: 0.0001
+- `learning_rate`: 8e-05
 - `weight_decay`: 0.005
-- `max_steps`: 562
+- `max_steps`: 500
 - `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`: 0.0001
+- `learning_rate`: 8e-05
 - `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`: 562
+- `max_steps`: 500
 - `lr_scheduler_type`: linear
 - `lr_scheduler_kwargs`: {}
 - `warmup_ratio`: 0.1
@@ -630,9 +630,9 @@ 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    | -             | 1.1445          | 0.5540                     | 0.5931                | 0.5735                       |
-| 0.2874 | 250  | 1.0487        | 0.8460          | 0.5005                     | 0.5178                | 0.5092                       |
-| 0.5747 | 500  | 0.9685        | 0.8172          | 0.5086                     | 0.4956                | 0.5021                       |
+| 0      | 0    | -             | 1.1142          | 0.5887                     | 0.5786                | 0.5836                       |
+| 0.2874 | 250  | 0.9921        | 0.8098          | 0.5442                     | 0.5263                | 0.5353                       |
+| 0.5747 | 500  | 0.915         | 0.7854          | 0.5389                     | 0.5087                | 0.5238                       |
 
 
 ### Framework Versions