Add new SentenceTransformer model

1_Pooling/config.json

@@ -0,0 +1,10 @@
+{
+  "word_embedding_dimension": 1024,
+  "pooling_mode_cls_token": true,
+  "pooling_mode_mean_tokens": false,
+  "pooling_mode_max_tokens": false,
+  "pooling_mode_mean_sqrt_len_tokens": false,
+  "pooling_mode_weightedmean_tokens": false,
+  "pooling_mode_lasttoken": false,
+  "include_prompt": true
+}

README.md

@@ -0,0 +1,695 @@
+---
+tags:
+- sentence-transformers
+- sentence-similarity
+- feature-extraction
+- generated_from_trainer
+- dataset_size:370
+- loss:MatryoshkaLoss
+- loss:MultipleNegativesRankingLoss
+base_model: Snowflake/snowflake-arctic-embed-l
+widget:
+- source_sentence: ' What is the proposed alternative to imposing random events purely
+    by chance on players?'
+  sentences:
+  - "half a day). \n• They are perceived to be new and innovative (despite being around\
+    \ since 1987). \n• They are easy to transport, requiring only pen and paper –\
+    \ with perhaps a few maps and \ncounters. \n• They work well in multi-domain,\
+    \ multi-agency contexts allowing all Actors to participate \nequally. \nA few\
+    \ Words of Warning \n• The fact that a Matrix Game requires little infrastructure\
+    \ can be a problem – it just \ndoesn't look sexy and the strengths that it can\
+    \ be done quickly with the minimum of \nfuss, can be reduced by efforts to make\
+    \ it look cool/expensive. \n• The non-quantitative nature of the game can frustrate\
+    \ analysts. \n• Matrix Games require an experience facilitator to run them."
+  - "inform the other players of their stated intentions. In many cases these are\
+    \ not really \n\"arguments\" as part of the game, so shouldn't count as their\
+    \ action for the turn, unless they \nwish to specify a measurable effect (such\
+    \ as increasing their approval ratings). \nTrade Agreements \nIn some games, trade\
+    \ forms a very important part of the game narrative. In most cases this \ncan\
+    \ be treated simply as part of the normal ebb and flow of the argument process.\
+    \ \nHowever, in some circumstances, particularly when timescales are long, trade\
+    \ can require \ngreater attention as to the nuances of the economic benefits and\
+    \ impacts. In these cases, it \nmay be necessary to get the two sides to make\
+    \ additional arguments as to what they expect"
+  - "possible throughout the game, having “random events” happen completely at random\
+    \ is \nproblematic. An Actor may be disadvantaged purely by chance, more than\
+    \ once during the \ngame, which can reduce their immersion and engagement. The\
+    \ narrative develops during \nthe game based on the decisions of the players and\
+    \ their reactions to the decisions of other \nplayers. Having random events imposed\
+    \ on them by chance breaks this “cause and effect” \ncycle and degrades the game\
+    \ flow. \nThe alternative is to give the random event to the participants. They\
+    \ will then make a \ndecision as to how this can contribute to the narrative being\
+    \ developed by the players. They"
+- source_sentence: ' What is the primary purpose of the game described in the context?'
+  sentences:
+  - "If you are using voting systems, either as Diceless Adjudication or as Estimative\
+    \ Probability, \nyou should take great care to ensure that the players are being\
+    \ as professional as possible, \nand not merely \"voting for themselves\" in a\
+    \ competitive manner. Many players can be quite \nvery competitive, so it may\
+    \ be necessary to not allow them to vote on their argument – and \nequally it\
+    \ may be necessary to keep an eye on players who are in direct competition. The\
+    \ \nintention is to develop a narrative, generating insights – rather than trying\
+    \ to win at all \ncosts. \n \n7 An example is https://www.turningtechnologies.eu/turningpoint/\
+    \ \n8 An example is https://www.polleverywhere.com/\n\fVersion 15 \nPage 14 of\
+    \ 52  \n© Tom Mouat 2019, 2020, 2022, 2023"
+  - "spend the time piling markers on counters. Tracks can be generic (in that they\
+    \ simply record \nthe number of plusses or minuses applied) or they might have\
+    \ specific \"trigger levels\" (in \nthat when the morale of the infantry is reduced\
+    \ to -3, the \"raw\" units will desert and return \nto their homes. \nIt can also\
+    \ be useful to have a \"Press\" actor whose job it is to record the results of\
+    \ \narguments (both visible to the public and those not), as well as putting the\
+    \ \"Press spin\" on \nthe events. This role can be useful in looking after the\
+    \ \"Consequence Management\" \nelements mentioned earlier. \nThe Components (and\
+    \ Characters) Affect the Game \nWhen participants are thinking on their feet,\
+    \ what they can see will affect what they argue"
+  - "materials, a short game, and small numbers of participants. If they want to conduct\
+    \ a \"deep \ndive\", this isn't the appropriate game - the purpose is to identify\
+    \ the insights – so make a \nnote and move on. The \"deep dive\" should follow\
+    \ later or in a different type of game. You \nshould, therefore, make sure you\
+    \ include this point in your introductory briefing so that the \nplayers are clear\
+    \ from the outset. \nWhen dealing with dominant people, who continually interrupt\
+    \ and dominate the \nArguments, you need to take a harder line. You should interrupt\
+    \ them when they interrupt \nanother player making a point. Point out to them\
+    \ that they had their chance. This isn't a"
+- source_sentence: ' Why should Big Projects or Long-Term Plans require no more than
+    three successful arguments in the game?'
+  sentences:
+  - "much on this single thing.  \nThis does not mean that arguments have to only\
+    \ be about things that can happen within \nthe turn length of the game. It is\
+    \ possible to make \"long term\" arguments like anything else. \nIf, in a Baltic\
+    \ game with week-long turns, you want to argue that an electricity cable \nbetween\
+    \ Sweden and Lithuania is to be built with the aim of reducing Lithuania's \n\
+    dependence on Russian energy, this would be judged as normal. It just would not\
+    \ come to \n \n9 I am indebted to Prof Rex Brynen for this suggestion.\n\fVersion\
+    \ 15 \nPage 23 of 52  \n© Tom Mouat 2019, 2020, 2022, 2023 \nfruition in the length\
+    \ of the game – but, assuming the argument was successful, it would"
+  - "games.\n\fVersion 15 \nPage 36 of 52  \n© Tom Mouat 2019, 2020, 2022, 2023 \n\
+    Why I like Matrix Games \n• Designing a Matrix Game can be done quickly with the\
+    \ minimum of fuss. \n• Participating in a Matrix Game does not require an understanding\
+    \ of complex and \nunfamiliar rules. \n• Matrix games can cover a wide variety\
+    \ of possible scenarios, including conceptual \nconflicts like Cyber. \n• They\
+    \ are especially good in the non-kinetic, effects based, domain. \n• Matrix games\
+    \ deal with qualitative outputs so are especially useful for non-analysts. \n\
+    • The games work best with small groups, increasing immersion and buy-in to the\
+    \ game. \n• Matrix games are extremely inexpensive (and they work best with short\
+    \ sessions lasting \nhalf a day)."
+  - "protection: Its hidden location, its boundary fence, and the security guards,\
+    \ all of which \nmust be overcome by successful arguments before the base can\
+    \ be penetrated. \nAs a rule of thumb, nothing should have more than 3 levels\
+    \ of protection as it will simply \ntake too long and dominate the game to the\
+    \ exclusion of everything else. \nBig Projects or Long-Term Plans \nDepending\
+    \ on the level of the game, some actions and events represent such a large \n\
+    investment in time and effort that they require multiple arguments in order to\
+    \ bring them \nto fruition. As a rule of thumb, a Big Project should also take\
+    \ no more than 3 successful \narguments (like protected and hidden things above);\
+    \ otherwise, the game is focussed too"
+- source_sentence: ' Which associations related to wargaming and simulation are mentioned
+    in the context?'
+  sentences:
+  - "out their objectives and explain why they though they succeeded or failed can\
+    \ be most \ninstructive. Also, if you then ask the assembled group \"who won?\"\
+    \ and they all agree, then \nthis can be a very powerful indicator of things that\
+    \ might need to be looked at more closely \nas a result of the game. \nFinally,\
+    \ the insights from the game can take a little time to come out. They might not\
+    \ be \nimmediately obvious, so taking time to consider what happened in the game\
+    \ and whether \nindividual events are noteworthy, is very useful. I am continually\
+    \ surprised at the predictive \npower of such a simple game. \n \n \n \n11 See:\
+    \ Game theory, simulated interaction, and unaided judgement for forecasting decisions\
+    \ in conflicts. Kesten C. Green."
+  - 'gaming vignettes
+
+
+    job opportunities/positions vacant
+
+
+    latest links
+
+
+    methodology
+
+
+    not-so-serious
+
+
+    playtesters needed
+
+
+    reader survey
+
+
+    request for proposals
+
+
+    scholarships and fellowships
+
+
+    simulation and game reports
+
+
+    simulation and game reviews
+
+
+    simulation and gaming debacles
+
+
+    simulation and gaming history
+
+
+    simulation and gaming ideas
+
+
+    simulation and gaming journals
+
+
+    simulation and gaming materials
+
+
+    simulation and gaming miscellany
+
+
+    simulation and gaming news
+
+
+    simulation and gaming publications
+
+
+    simulation and gaming software
+
+
+    Archives
+
+
+    M T W T F S S 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
+    26 27 28 29 30
+
+
+    Associations
+
+
+    Australian Defence Force Wargaming Group
+
+
+    Connections Netherlands
+
+
+    Connections North (Canada)'
+  - "Senior Officers, Dominant People and Contentious Arguments \nIt is not uncommon\
+    \ in a Matrix Game that the participants want to \"debate\" the arguments. \n\
+    To a limited extent this is ok, but as stated elsewhere, the game needs to move\
+    \ at a pace, \ncreating an immersive narrative and forcing the players to have\
+    \ to live with the \nconsequences of their earlier decisions.  \nIt can happen\
+    \ that a Senior Officer, used to \"seminar wargames\", will interrupt when you\
+    \ \nwant to move on and say \"wait a minute - this is a really valuable debate\
+    \ - let's just dig \ndown...\" You should try to point out that this is not that\
+    \ sort of game - Matrix Games are to \ngain an insight and understanding in a\
+    \ specific way. Short notice, minimal preparation and"
+- source_sentence: ' Why is it important for player roles in a Matrix Game to operate
+    at broadly similar levels?'
+  sentences:
+  - 'The Basic Rule.   The basic rule is as follows: 1 x 6-Sided Dice = 1 x Combat
+    Unit The size of that Combat Unit will, of course, vary from game to game. In
+    the boarding action it may be as little as 5-10 men; in a Map Game, it could be
+    as much as an entire Brigade, or even a Corps.
+
+
+    The Method.   The dice on the opposing sides are rolled as follows: Roll the Dice.
+    Line them up, Highest vs Highest If one side has more dice than the other, any
+    dice that are extra, and score less than the lowest dice of the side with the
+    fewer dice, are ignored.'
+  - "Matrix Game Checklist ....................................................................................\
+    \ 38 \nSample Spendable Bonus Cards ......................................................................\
+    \ 40 \nSample Random Events ...................................................................................\
+    \ 41 \nSample Voting Cards for Diceless Adjudication ...............................................\
+    \ 43 \nSample Estimative Probability Cards ...............................................................\
+    \ 44 \nSample Turn Order Cards ................................................................................\
+    \ 45 \nSample Markers for Matrix Games for Effects and Conventional Forces ........\
+    \ 46"
+  - 'When you are designing a Matrix Game it is worth thinking about the level at
+    which the players roles will be operating in the game. In is usually better, and
+    produces a more balanced game, when the level on which the player roles are operating
+    are broadly similar. It would be difficult to get a balanced game if 3 of the
+    players are playing Generals in command of vast Armies, and another player is
+    playing a simple individual soldier.
+
+
+    Levels of Protection and Hidden Things.'
+pipeline_tag: sentence-similarity
+library_name: sentence-transformers
+metrics:
+- cosine_accuracy@1
+- cosine_accuracy@3
+- cosine_accuracy@5
+- cosine_accuracy@10
+- cosine_precision@1
+- cosine_precision@3
+- cosine_precision@5
+- cosine_precision@10
+- cosine_recall@1
+- cosine_recall@3
+- cosine_recall@5
+- cosine_recall@10
+- cosine_ndcg@10
+- cosine_mrr@10
+- cosine_map@100
+model-index:
+- name: SentenceTransformer based on Snowflake/snowflake-arctic-embed-l
+  results:
+  - task:
+      type: information-retrieval
+      name: Information Retrieval
+    dataset:
+      name: Unknown
+      type: unknown
+    metrics:
+    - type: cosine_accuracy@1
+      value: 0.9347826086956522
+      name: Cosine Accuracy@1
+    - type: cosine_accuracy@3
+      value: 1.0
+      name: Cosine Accuracy@3
+    - type: cosine_accuracy@5
+      value: 1.0
+      name: Cosine Accuracy@5
+    - type: cosine_accuracy@10
+      value: 1.0
+      name: Cosine Accuracy@10
+    - type: cosine_precision@1
+      value: 0.9347826086956522
+      name: Cosine Precision@1
+    - type: cosine_precision@3
+      value: 0.33333333333333337
+      name: Cosine Precision@3
+    - type: cosine_precision@5
+      value: 0.1999999999999999
+      name: Cosine Precision@5
+    - type: cosine_precision@10
+      value: 0.09999999999999995
+      name: Cosine Precision@10
+    - type: cosine_recall@1
+      value: 0.9347826086956522
+      name: Cosine Recall@1
+    - type: cosine_recall@3
+      value: 1.0
+      name: Cosine Recall@3
+    - type: cosine_recall@5
+      value: 1.0
+      name: Cosine Recall@5
+    - type: cosine_recall@10
+      value: 1.0
+      name: Cosine Recall@10
+    - type: cosine_ndcg@10
+      value: 0.97023760333851
+      name: Cosine Ndcg@10
+    - type: cosine_mrr@10
+      value: 0.9601449275362318
+      name: Cosine Mrr@10
+    - type: cosine_map@100
+      value: 0.960144927536232
+      name: Cosine Map@100
+---
+
+# SentenceTransformer based on Snowflake/snowflake-arctic-embed-l
+
+This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [Snowflake/snowflake-arctic-embed-l](https://huggingface.co/Snowflake/snowflake-arctic-embed-l). It maps sentences & paragraphs to a 1024-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:** [Snowflake/snowflake-arctic-embed-l](https://huggingface.co/Snowflake/snowflake-arctic-embed-l) <!-- at revision d8fb21ca8d905d2832ee8b96c894d3298964346b -->
+- **Maximum Sequence Length:** 512 tokens
+- **Output Dimensionality:** 1024 dimensions
+- **Similarity Function:** Cosine Similarity
+<!-- - **Training Dataset:** Unknown -->
+<!-- - **Language:** Unknown -->
+<!-- - **License:** Unknown -->
+
+### Model Sources
+
+- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
+- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
+- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
+
+### Full Model Architecture
+
+```
+SentenceTransformer(
+  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
+  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
+  (2): Normalize()
+)
+```
+
+## Usage
+
+### Direct Usage (Sentence Transformers)
+
+First install the Sentence Transformers library:
+
+```bash
+pip install -U sentence-transformers
+```
+
+Then you can load this model and run inference.
+```python
+from sentence_transformers import SentenceTransformer
+
+# Download from the 🤗 Hub
+model = SentenceTransformer("CalebMaresca/matrix-game-embeddings-ft-v1")
+# Run inference
+sentences = [
+    ' Why is it important for player roles in a Matrix Game to operate at broadly similar levels?',
+    'When you are designing a Matrix Game it is worth thinking about the level at which the players roles will be operating in the game. In is usually better, and produces a more balanced game, when the level on which the player roles are operating are broadly similar. It would be difficult to get a balanced game if 3 of the players are playing Generals in command of vast Armies, and another player is playing a simple individual soldier.\n\nLevels of Protection and Hidden Things.',
+    'Matrix Game Checklist .................................................................................... 38 \nSample Spendable Bonus Cards ...................................................................... 40 \nSample Random Events ................................................................................... 41 \nSample Voting Cards for Diceless Adjudication ............................................... 43 \nSample Estimative Probability Cards ............................................................... 44 \nSample Turn Order Cards ................................................................................ 45 \nSample Markers for Matrix Games for Effects and Conventional Forces ........ 46',
+]
+embeddings = model.encode(sentences)
+print(embeddings.shape)
+# [3, 1024]
+
+# Get the similarity scores for the embeddings
+similarities = model.similarity(embeddings, embeddings)
+print(similarities.shape)
+# [3, 3]
+```
+
+<!--
+### Direct Usage (Transformers)
+
+<details><summary>Click to see the direct usage in Transformers</summary>
+
+</details>
+-->
+
+<!--
+### Downstream Usage (Sentence Transformers)
+
+You can finetune this model on your own dataset.
+
+<details><summary>Click to expand</summary>
+
+</details>
+-->
+
+<!--
+### Out-of-Scope Use
+
+*List how the model may foreseeably be misused and address what users ought not to do with the model.*
+-->
+
+## Evaluation
+
+### Metrics
+
+#### Information Retrieval
+
+* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
+
+| Metric              | Value      |
+|:--------------------|:-----------|
+| cosine_accuracy@1   | 0.9348     |
+| cosine_accuracy@3   | 1.0        |
+| cosine_accuracy@5   | 1.0        |
+| cosine_accuracy@10  | 1.0        |
+| cosine_precision@1  | 0.9348     |
+| cosine_precision@3  | 0.3333     |
+| cosine_precision@5  | 0.2        |
+| cosine_precision@10 | 0.1        |
+| cosine_recall@1     | 0.9348     |
+| cosine_recall@3     | 1.0        |
+| cosine_recall@5     | 1.0        |
+| cosine_recall@10    | 1.0        |
+| **cosine_ndcg@10**  | **0.9702** |
+| cosine_mrr@10       | 0.9601     |
+| cosine_map@100      | 0.9601     |
+
+<!--
+## Bias, Risks and Limitations
+
+*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
+-->
+
+<!--
+### Recommendations
+
+*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
+-->
+
+## Training Details
+
+### Training Dataset
+
+#### Unnamed Dataset
+
+* Size: 370 training samples
+* Columns: <code>sentence_0</code> and <code>sentence_1</code>
+* Approximate statistics based on the first 370 samples:
+  |         | sentence_0                                                                         | sentence_1                                                                          |
+  |:--------|:-----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|
+  | type    | string                                                                             | string                                                                              |
+  | details | <ul><li>min: 11 tokens</li><li>mean: 20.19 tokens</li><li>max: 34 tokens</li></ul> | <ul><li>min: 8 tokens</li><li>mean: 150.83 tokens</li><li>max: 512 tokens</li></ul> |
+* Samples:
+  | sentence_0                                                                                                                                            | sentence_1                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                     |
+  |:------------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+  | <code> What distinguishes "established facts" from other types of facts in the game briefings or play?</code>                                         | <code>Forces soldiers are going to be much more effective in combat than untrained protestors; <br>and "established facts" which are facts that have been specifically mentioned in the game <br>briefings or have become established during play as the result of successful arguments.  <br>The latter can be immediately deployed as supporting reasons (Pros and Cons), but the <br>former need to have been argued successfully in order for them to be specifically included. <br>Many inexperienced players will make vast all-encompassing arguments full of assumptions <br>that are not reasonable. For example: It is not a reasonable assumption that unarmed <br>Protestors could fight off trained Police. It is reasonable to assume that the Police are</code> |
+  | <code> Why is it unreasonable to assume that unarmed protestors could fight off trained police according to the context?</code>                       | <code>Forces soldiers are going to be much more effective in combat than untrained protestors; <br>and "established facts" which are facts that have been specifically mentioned in the game <br>briefings or have become established during play as the result of successful arguments.  <br>The latter can be immediately deployed as supporting reasons (Pros and Cons), but the <br>former need to have been argued successfully in order for them to be specifically included. <br>Many inexperienced players will make vast all-encompassing arguments full of assumptions <br>that are not reasonable. For example: It is not a reasonable assumption that unarmed <br>Protestors could fight off trained Police. It is reasonable to assume that the Police are</code> |
+  | <code> What was the outcome of the initial Russian attack against the German units, and how did it affect the ammunition status of both sides?</code> | <code>The Russians succeed in pushing back one of the German units and forcing and already depleted unit to use up ammunition, (but are pushed back themselves and 2 units use a lot of ammo (one of which becomes combat ineffective on -3)). Overall, as the success is matched by failure, the line itself holds. The Russians attack again, the next day:<br><br>Initial Dice Throw:		RUSSIAN:	6 5 5 4 2 4 GERMAN:		1 2 4 4 Lined Up and Modified:	RUSSIAN:	5 4 3 3 2 1		(two of the Russians = -2) GERMAN:		4 3 3 2			(one of the Germans = -1) Result of Third Day;	lose:						(one of the Germans = +0) RUSSIAN:	5 4 3 3 GERMAN:	    	3 lose:		4 3   2</code>                                                                                                           |
+* Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
+  ```json
+  {
+      "loss": "MultipleNegativesRankingLoss",
+      "matryoshka_dims": [
+          768,
+          512,
+          256,
+          128,
+          64
+      ],
+      "matryoshka_weights": [
+          1,
+          1,
+          1,
+          1,
+          1
+      ],
+      "n_dims_per_step": -1
+  }
+  ```
+
+### Training Hyperparameters
+#### Non-Default Hyperparameters
+
+- `eval_strategy`: steps
+- `per_device_train_batch_size`: 10
+- `per_device_eval_batch_size`: 10
+- `num_train_epochs`: 10
+- `multi_dataset_batch_sampler`: round_robin
+
+#### All Hyperparameters
+<details><summary>Click to expand</summary>
+
+- `overwrite_output_dir`: False
+- `do_predict`: False
+- `eval_strategy`: steps
+- `prediction_loss_only`: True
+- `per_device_train_batch_size`: 10
+- `per_device_eval_batch_size`: 10
+- `per_gpu_train_batch_size`: None
+- `per_gpu_eval_batch_size`: None
+- `gradient_accumulation_steps`: 1
+- `eval_accumulation_steps`: None
+- `torch_empty_cache_steps`: None
+- `learning_rate`: 5e-05
+- `weight_decay`: 0.0
+- `adam_beta1`: 0.9
+- `adam_beta2`: 0.999
+- `adam_epsilon`: 1e-08
+- `max_grad_norm`: 1
+- `num_train_epochs`: 10
+- `max_steps`: -1
+- `lr_scheduler_type`: linear
+- `lr_scheduler_kwargs`: {}
+- `warmup_ratio`: 0.0
+- `warmup_steps`: 0
+- `log_level`: passive
+- `log_level_replica`: warning
+- `log_on_each_node`: True
+- `logging_nan_inf_filter`: True
+- `save_safetensors`: True
+- `save_on_each_node`: False
+- `save_only_model`: False
+- `restore_callback_states_from_checkpoint`: False
+- `no_cuda`: False
+- `use_cpu`: False
+- `use_mps_device`: False
+- `seed`: 42
+- `data_seed`: None
+- `jit_mode_eval`: False
+- `use_ipex`: False
+- `bf16`: False
+- `fp16`: False
+- `fp16_opt_level`: O1
+- `half_precision_backend`: auto
+- `bf16_full_eval`: False
+- `fp16_full_eval`: False
+- `tf32`: None
+- `local_rank`: 0
+- `ddp_backend`: None
+- `tpu_num_cores`: None
+- `tpu_metrics_debug`: False
+- `debug`: []
+- `dataloader_drop_last`: False
+- `dataloader_num_workers`: 0
+- `dataloader_prefetch_factor`: None
+- `past_index`: -1
+- `disable_tqdm`: False
+- `remove_unused_columns`: True
+- `label_names`: None
+- `load_best_model_at_end`: False
+- `ignore_data_skip`: False
+- `fsdp`: []
+- `fsdp_min_num_params`: 0
+- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
+- `tp_size`: 0
+- `fsdp_transformer_layer_cls_to_wrap`: None
+- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
+- `deepspeed`: None
+- `label_smoothing_factor`: 0.0
+- `optim`: adamw_torch
+- `optim_args`: None
+- `adafactor`: False
+- `group_by_length`: False
+- `length_column_name`: length
+- `ddp_find_unused_parameters`: None
+- `ddp_bucket_cap_mb`: None
+- `ddp_broadcast_buffers`: False
+- `dataloader_pin_memory`: True
+- `dataloader_persistent_workers`: False
+- `skip_memory_metrics`: True
+- `use_legacy_prediction_loop`: False
+- `push_to_hub`: False
+- `resume_from_checkpoint`: None
+- `hub_model_id`: None
+- `hub_strategy`: every_save
+- `hub_private_repo`: None
+- `hub_always_push`: False
+- `gradient_checkpointing`: False
+- `gradient_checkpointing_kwargs`: None
+- `include_inputs_for_metrics`: False
+- `include_for_metrics`: []
+- `eval_do_concat_batches`: True
+- `fp16_backend`: auto
+- `push_to_hub_model_id`: None
+- `push_to_hub_organization`: None
+- `mp_parameters`: 
+- `auto_find_batch_size`: False
+- `full_determinism`: False
+- `torchdynamo`: None
+- `ray_scope`: last
+- `ddp_timeout`: 1800
+- `torch_compile`: False
+- `torch_compile_backend`: None
+- `torch_compile_mode`: None
+- `include_tokens_per_second`: False
+- `include_num_input_tokens_seen`: False
+- `neftune_noise_alpha`: None
+- `optim_target_modules`: None
+- `batch_eval_metrics`: False
+- `eval_on_start`: False
+- `use_liger_kernel`: False
+- `eval_use_gather_object`: False
+- `average_tokens_across_devices`: False
+- `prompts`: None
+- `batch_sampler`: batch_sampler
+- `multi_dataset_batch_sampler`: round_robin
+
+</details>
+
+### Training Logs
+| Epoch  | Step | cosine_ndcg@10 |
+|:------:|:----:|:--------------:|
+| 1.0    | 37   | 0.9273         |
+| 1.3514 | 50   | 0.9490         |
+| 2.0    | 74   | 0.9462         |
+| 2.7027 | 100  | 0.9527         |
+| 3.0    | 111  | 0.9527         |
+| 4.0    | 148  | 0.9783         |
+| 4.0541 | 150  | 0.9811         |
+| 5.0    | 185  | 0.9622         |
+| 5.4054 | 200  | 0.9622         |
+| 6.0    | 222  | 0.9702         |
+| 6.7568 | 250  | 0.9622         |
+| 7.0    | 259  | 0.9622         |
+| 8.0    | 296  | 0.9702         |
+| 8.1081 | 300  | 0.9702         |
+| 9.0    | 333  | 0.9702         |
+| 9.4595 | 350  | 0.9702         |
+| 10.0   | 370  | 0.9702         |
+
+
+### Framework Versions
+- Python: 3.13.2
+- Sentence Transformers: 4.1.0
+- Transformers: 4.51.3
+- PyTorch: 2.7.0+cu126
+- Accelerate: 1.6.0
+- Datasets: 3.6.0
+- Tokenizers: 0.21.1
+
+## Citation
+
+### BibTeX
+
+#### Sentence Transformers
+```bibtex
+@inproceedings{reimers-2019-sentence-bert,
+    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
+    author = "Reimers, Nils and Gurevych, Iryna",
+    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
+    month = "11",
+    year = "2019",
+    publisher = "Association for Computational Linguistics",
+    url = "https://arxiv.org/abs/1908.10084",
+}
+```
+
+#### MatryoshkaLoss
+```bibtex
+@misc{kusupati2024matryoshka,
+    title={Matryoshka Representation Learning},
+    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
+    year={2024},
+    eprint={2205.13147},
+    archivePrefix={arXiv},
+    primaryClass={cs.LG}
+}
+```
+
+#### MultipleNegativesRankingLoss
+```bibtex
+@misc{henderson2017efficient,
+    title={Efficient Natural Language Response Suggestion for Smart Reply},
+    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
+    year={2017},
+    eprint={1705.00652},
+    archivePrefix={arXiv},
+    primaryClass={cs.CL}
+}
+```
+
+<!--
+## Glossary
+
+*Clearly define terms in order to be accessible across audiences.*
+-->
+
+<!--
+## Model Card Authors
+
+*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
+-->
+
+<!--
+## Model Card Contact
+
+*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
+-->

config.json

@@ -0,0 +1,24 @@
+{
+  "architectures": [
+    "BertModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.51.3",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

config_sentence_transformers.json

@@ -0,0 +1,12 @@
+{
+  "__version__": {
+    "sentence_transformers": "4.1.0",
+    "transformers": "4.51.3",
+    "pytorch": "2.7.0+cu126"
+  },
+  "prompts": {
+    "query": "Represent this sentence for searching relevant passages: "
+  },
+  "default_prompt_name": null,
+  "similarity_fn_name": "cosine"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e02936f513c16a2ab95b17470cd43bac1ed9f3252c45ce6a7a05259c3dbea3aa
+size 1336413848

modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+  "max_seq_length": 512,
+  "do_lower_case": false
+}

special_tokens_map.json

@@ -0,0 +1,37 @@
+{
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,63 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "max_length": 512,
+  "model_max_length": 512,
+  "pad_to_multiple_of": null,
+  "pad_token": "[PAD]",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "sep_token": "[SEP]",
+  "stride": 0,
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "[UNK]"
+}