steamfit-encoder

SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2

This is a sentence-transformers model finetuned from sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 on the parquet dataset. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for retrieval.

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 <!-- at revision e8f8c211226b894fcb81acc59f3b34ba3efd5f42 -->
• Maximum Sequence Length: 192 tokens
• Output Dimensionality: 384 dimensions
• Similarity Function: Cosine Similarity
• Supported Modality: Text
• Training Dataset:
  • parquet

<!-- - **Language:** Unknown --> <!-- - **License:** Unknown -->

Model Sources

• Documentation: Sentence Transformers Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'BertModel'})
  (1): Pooling({'embedding_dimension': 384, 'pooling_mode': 'mean', 'include_prompt': True})
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
    '混世萌王喵霸霸 Cute Chaos Demon Nyanbaba | Funny | Casual | Cute | Comedy | Female Protagonist | Hand-drawn | Investigation | Point & Click | Exploration | Indie | Incremental | Hidden Object | Anime | Creature Collector | Minimalist | Psychological Horror | Family Friendly | Puzzle | Horror | Logic | Casual | Indie | The super cute villain Nyanbaba is back again! This time, meow bully invaded 10 new scenes, found all the strange hidden points and cracked the magic of cat ear Nyanbaba! Contains jump out shock (cute)! And super cute Chinese dubbing! More scenes, more voice, more joy!  ‖ 사용자 리뷰: 냥바바 2탄! 퍼즐은 여전히 어렵다. 귀여운 일러랑 도전과제 줍줍하기 좋아요 :)  ·  we love cheap hidden object games  ·  short and sweet, challenging but not so challenging it\'ll take you a year to beat it! i recommend it if you just love hidden item games but don\'t want the commitment of one with hundreds of levels.  ·  Another good Nyanbaba game! This one is easier, but the artworks are interesting and look great. For those who don’t know, Nyanbaba is a hidden-object puzzle game that is different by one characteristic: you have to find  ·  very cute!! and fun! :))  ·  i love this whole series lol  ·  可爱滴捏 く__,.ヘヽ.\u3000\u3000\u3000\u3000/\u3000,ー､ 〉 \u3000\u3000\u3000\u3000\u3000＼ \', !-─‐-i\u3000/\u3000/´ \u3000\u3000\u3000 \u3000 ／｀ｰ\'\u3000\u3000\u3000 L/／｀ヽ､ \u3000\u3000 \u3000 /\u3000 ／,\u3000 /|\u3000 ,\u3000 ,\u3000\u3000\u3000 \', \u3000\u3000\u3000ｲ \u3000/ /-‐/\u3000ｉ\u3000L_ ﾊ ヽ!\u3000 i \u3000\u3000\u3000 ﾚ ﾍ 7ｲ｀ﾄ\u3000 ﾚ\'ｧ-ﾄ､!ハ|\u3000 | \u3000\u3000\u3000\u3000 !,/7 \'0\'\u3000\u3000 ´0iソ| \u3000 |\u3000\u3000\u3000 \u3000\u3000\u3000\u3000 |.从"\u3000\u3000_\u3000\u3000 ,,,, / |./ \u3000 | \u3000\u3000\u3000\u3000 ﾚ\'| i＞.､  ·  黑妖的喵霸霸系列找茬游戏最新作，玩家将在10个不同的场景中各自寻找10处氛围不同的可疑区域，破除猫娘的幻术。如果在限定时间内没能成功，将会收获可爱主角的Q萌冲击。和前作一样，秘籍6与7可以分别提前解锁二周目迷雾和增加至20处隐藏点。找茬途中和触发通关或败北时均有配音，想要全成就也必须历经全部关卡通关、二周目通关以及败北。喜欢前作的玩家不要错过！  [quote] [b][u] ❀书以一家之言🌸而荐一人之好❀ [/u][/b]  ❀欢迎→  ·  依旧是60分钟局，不多不少刚刚好 依旧是这么萌的画风和配音 为可爱干杯！ 期待下一作',
    "Mosaique Neko Waifus 3 | Hentai | Puzzle | RPG | Cute | Visual Novel | Interactive Fiction | Collectathon | Dating Sim | Anime | Romance | 2D | Colorful | Hand-drawn | Casual | LGBTQ+ | Choices Matter | Inventory Management | Nudity | Sexual Content | CRPG | Casual | RPG | Anime girl puzzle - RPG - micro visual novel. 4 Cute neko girls want to know all about you. Puzzles, RPG, Collectible items that boost your skills!  ‖ 사용자 리뷰: [h1]최고의 가성비로 Live2D꼴림가득 꼭띠와 뷰지를 볼수있다[/h1] [table]     [tr]         [th]가성비[/th]         [th]10점[/th]     [/tr]     [tr]         [td]뷰지[/td]         [td]O[/td]     [/tr]     [tr]         [td]꼭띠[/td]         [td]O[/td]      ·  R18 무료 DLC를 같이 다운받고 하자. 일단 전작보다 그림체가 좋아진 느낌이다. 전작처럼 다양한 스킬같은게있는데 이펙트는 화려하지만 정도가 너무 지나쳐 눈이 아플정도이다. 하지만 가격도 저렴하기때문에 추천한다.  ·  세일 : -80% ₩ 680  트레이딩 카드 농사용 게임.  ·  최초이차 최후로 남녀합일 운우지정을 묘사한 게임으로 시리즈 최고 명작이 아닐까 한다.  4편이랑 5편도 나쁜 건 아니지만 어차피 게임 플레이는 그게 그거인데 남녀합일은 기대도 안 했지만 2까지 넣어주던 여성끼리 우정을 다지는 장면도 4부턴 없어져서 네코 3이 제일 고점이 아닌가 싶다.  솔직히 1,2 대비 그 외 인게임이 크게 달라진 건 없는데 편의성이 많이 좋아졌다.  다만 단점이 있다면   ·  게임 너무 잘 만든듯  재밌게 플레이 했어요 감사합니다 ~!  ·  - 졸업 - (100%)  게임 정복 쉬움 치트키 안쓰고 했음  ·  𝙈𝙤𝙨𝙖𝙞𝙦𝙪𝙚 𝙉𝙚𝙠𝙤 𝙒𝙖𝙞𝙛𝙪𝙨 𝙖𝙧𝙚 𝙩𝙝𝙚 𝙗𝙚𝙨𝙩 𝙩𝙞𝙢𝙚𝙨 𝟯~  ·  ⣿⣿⣿⣿⣿⣿⣿⡇⡌⡰⢃⡿⡡⠟⣠⢹⡏⣦⢸⣿⣿⣿⣿⣿⣿ ⣿⣿⣿⣿⣿⣿⡿⢰⠋⡿⢋⣐⡈⣽⠟⢀⢻⢸⡂⣿⣿⣿⣿⣿⣿ ⣿⣿⣿⣿⣿⣋⠴⢋⡘⢰⣄⣀⣅⣡⠌⠛⠆⣿⡄⣿⣿⣿⣿⣿⣿ ⣿⣿⣿⣿⣿⣿⣶⣁⣐⠄⠹⣟⠯⢿⣷⠾⠁⠥⠃⣹⣿⣿⣿⣿⣿ ⣿⣿⣿⣿⠟⠋⡍⢴⣶⣶⣶⣤⣭⡐⢶⣾⣿⣶⡆⢨⠛⠻⣿⣿⣿ ⣿⣿⣿⢏⣘⣚⣣⣾⣿⣿⣿⣿⣿⣿⢈⣿⣿⣿⣧⣘⠶⢂⠹⣿⣿ ⣿⣿⠃⣾⣿⣿⣿⣿⣿⣿⡿⠿⠿⠿⡀⢿⣿⣿⣿⣿⣿⣿⡇⣿⣿ ⣿⣿⡄⣿⣿⣿⣿⣿⣿⡯⠄⠄⠾⠿⠿⢦⣝⠻⣿⣿⣿⣿⠇⣿⣿ ⣿⣿⣷⣜⠿⢿⣿⡿⠟⣴⣾⣿  ·  This picture below is more hard to make than this game  ⣿⣿⡻⠿⣳⠸⢿⡇⢇⣿⡧⢹⠿⣿⣿⣿⣿⣾⣿⡇⣿⣿⣿⣿⡿⡐⣯⠁ ⠄⠄ ⠟⣛⣽⡳⠼⠄⠈⣷⡾⣥⣱⠃⠣⣿⣿⣿⣯⣭⠽⡇⣿⣿⣿⣿⣟⢢⠏⠄ ⠄ ⢠⡿⠶⣮⣝⣿⠄⠄⠈⡥⢭⣥⠅⢌⣽⣿⣻⢶⣭⡿⠿⠜⢿⣿⣿⡿⠁⠄⠄ ⠄⣼⣧⠤⢌⣭⡇⠄⠄⠄⠭⠭⠭⠯⠴⣚⣉⣛⡢⠭⠵⢶⣾⣦⡍⠁⠄⠄⠄⠄ ⠄⣿⣷⣯⣭⡷⠄⠄⢀⣀⠩⠍⢉⣛⣛⠫⢏⣈⣭⣥⣶⣶⣦⣭⣛⠄⠄⠄⠄⠄ ⢀⣿⣿⣿⡿⠃⢀⣴⣿  ·  nooo onii-saaan~ ⣿⣿⣷⡁⢆⠈⠕⢕⢂⢕⢂⢕⢂⢔⢂⢕⢄⠂⣂⠂⠆⢂⢕⢂⢕⢂⢕⢂⢕⢂ ⣿⣿⣿⡷⠊⡢⡹⣦⡑⢂⢕⢂⢕⢂⢕⢂⠕⠔⠌⠝⠛⠶⠶⢶⣦⣄⢂⢕⢂⢕ ⣿⣿⠏⣠⣾⣦⡐⢌⢿⣷⣦⣅⡑⠕⠡⠐⢿⠿⣛⠟⠛⠛⠛⠛⠡⢷⡈⢂⢕⢂ ⠟⣡⣾⣿⣿⣿⣿⣦⣑⠝⢿⣿⣿⣿⣿⣿⡵⢁⣤⣶⣶⣿⢿⢿⢿⡟⢻⣤⢑⢂ ⣾⣿⣿⡿⢟⣛⣻⣿⣿⣿⣦⣬⣙⣻⣿⣿⣷⣿⣿⢟⢝⢕⢕⢕⢕⢽⣿⣿⣷⣔ ⣿⣿⠵⠚⠉⢀⣀⣀⣈⣿⣿⣿⣿⣿⣿⣿⣿⣿⣗⢕⢕⢕⢕⢕⢕⣽⣿⣿⣿⣿ ⢷⣂⣠⣴⣾⡿⡿⡻⡻⣿⣿⣴⣿⣿⣿⣿⣿  ·  [b]I play for touching story and immersive gameplay. You play for bobas. We are not the same.[/b] ⣼⣿⣿⢿⡻⢝⠙⠊⠋⠉⠉⠈⠊⠝⣿⡻⠫⠫⠊⠑⠉⠉⠑⠫⢕⡫⣕⡁⠁ ⣼⡻⠕⠅⠁⣀⣤⣤⣄⣀⠈⠄⠁⠄⠁⣿⡮⠄⠁⠄⠄⡠⠶⠶⠦⡀⠈⣽⡢ ⣿⣧⠄⠁⠄⠔⠒⠭⠭⠥⠥⠓⠄⢀⣴⣿⣿⡄⠁⠠⣤⠉⠉⣭⠝⠈⢐⣽⣕ ⣿⣷⡢⢄⡰⡢⡙⠄⠠⠛⠁⢀⢔⣵⣿⣿⣿⣿⣧⣄⡈⠁⠈⠁⠉⡹⣽⣿⣷ ⣿⣿  ·  要是看的每一条缩牛子评论都显灵的话，老子牛子都快缩到脑子下面了。你发这个就是信这个是吧，点赞这条评论，每个发缩牛子评论的牛子缩短1cm，点赞的身体健康，你的牛子已经很完美了，不需要其他外力，请爱护它。  ·  今天是牛子节，点赞这段话，牛子精灵就会在你睡着的时候偷偷给你的牛子加长1cm，但是如果你看到了这段话不点赞的话，你的牛子就会缩短1cm    __..,,__\u3000\u3000\u3000,.｡='`1 \u3000\u3000\u3000\u3000 .,,..;~`''''\u3000\u3000\u3000\u3000`''''＜``彡\u3000} \u3000 _...:=,`'\u3000\u3000 \u3000︵\u3000 т\u3000︵\u3000\u3000X彡-J ＜`\u3000彡 /\u3000\u3000ミ\u3000\u3000,_人_.\u3000＊彡\u3000`~ \u3000 `~=::\u3000\u3000\u3000 \u3000\u3000\u3000\u3000\u3000\u3000 \u3000\u3000\u3000Y \u3000\u3000 \u3000i.\u3000\u3000\u3000\u3000\u3000\u3000\u3000\u3000\u3000\u3000\u3000\u3000 .: \u3000  ·  传递爱心: steam://install/1389910",
    'Curse of the Dead Gods | Action Roguelike | Isometric | Dark Fantasy | Roguelite | Difficult | Action | Roguelike | Souls-like | Fighting | Dungeon Crawler | Adventure | Singleplayer | Indie | Exploration | Real-Time | Great Soundtrack | 3D | Cartoony | Fantasy | Horror | Action | Adventure | You seek untold riches, eternal life, divine powers - it leads to this accursed temple, a seemingly-infinite labyrinth of bottomless pits, deadly traps, and monsters.  ‖ 사용자 리뷰: 타격감 좋고 재미는 있는데 난이도 조절을 어거지로 병신같이 해서 불쾌하게 어려움  ·  하데스의 빠릿빠릿한 액션을 기대했으면 실망함 대신 묵직한 한방과 패링 손맛은 일품 빛과 어둠을 활용한 시스템은 참신함 근데 뭐만하면 꺼져버리는 횃불 때문에 불편하고 내내 캄캄한 화면에서 전투해야 하는건 답답함 어렵고 성취감 있는 게임을 즐겨한다면 괜찮은 선택임  ·  이,,오,락,,,아주,,,진국,입,니다,,! 저,,같이,,나이,,많은,,오,락즐기미두,,,무난허게,,,잘,,할,수,,있구,,,  패-링,과,회피의,,조화,,,활력의,안배,를,,잘,,생각하며,,전략적인,,오락을,하다,보니,,,    저두,모르게,,,제갈량이,,되고,,있었읍니다,,,,~~!!~!~!!엄지,,,,척,,,~!~!~!~!!  ·  반복 플레이를 너무 많이 요구함, 하지만 패링이랑 회피가 즐겁다면 견딜만함  ·  게임은 어렵고 재미있다.  재미가 없다는 사람은 포기하고 도망가서 그런거다.    ·  할인 할때 구입하면 싼마이한 값에 즐기기 괜찮음  ·  in an alternate universe this game would have gotten as many accolades as Hades (in terms of its gameplay loop)  ·  Is there anything to explore? not really  Does it have a story? barely    If fast paced hard combat is your thing, this game is right for you. There is enough variety in terms of enemies and bosses. You will fight only a  ·  I like the combat better than Hades - there I said it.  The parry system is a nice twist and combat is meatier.  Also more weapons and effects than Hades.  Hades is better in terms of environment, storytelling, art style  ·  The best combat system in the genre. The game needs a sequel.  ·  Great game. Enjoyable difficulty ramp. Parries are satisfactory. Good Art. Maybe more unlockables would be great, other than that, 9/10  ·  更新：收回我之前的评测，作为一个打折后20元以下的游戏来说，它的性价比是极高的。各种打击感和UI的手感都是一流，画面也很有质感。其实还是有很高的可玩性的。  但是与此同时，怪物设计、流程设计的恶心也是极具代表性的，频繁一打多、霸体、弹反、无敌、招小怪……这些游戏战斗中典型的恶心设计在游戏中比比皆是，时刻考验你的耐心  ---  食之无味，弃之可惜。  如果说《哈迪斯》给人带来的是一场绝妙的艺术体验，让人沉浸于北欧神话的氛围之中。而《空洞  ·  这个制作组总是能做出一款很不错的恶心游戏······这是怎么做到的？  ·  这款游戏在“动作性”上是拔群的，一只手数的过来的肉鸽在动作性好的那种拔群  不玩可惜。',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.7986, 0.6517],
#         [0.7986, 1.0000, 0.5981],
#         [0.6517, 0.5981, 1.0000]])

<!-- ### 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.* --> <!-- ## 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

parquet

• Dataset: parquet
• Size: 1,500,000 training samples
• Columns: <code>anchor</code> and <code>positive</code>
• Approximate statistics based on the first 100 samples:

  	anchor	positive
  type	string	string
  modality	text	text
  details	<ul><li>min: 192 tokens</li><li>mean: 192.0 tokens</li><li>max: 192 tokens</li></ul>	<ul><li>min: 192 tokens</li><li>mean: 192.0 tokens</li><li>max: 192 tokens</li></ul>

• Samples:

  anchor	positive
  <code>Thief Simulator | Simulation | Stealth | Singleplayer | Crime | First-Person | Co-op | Open World | Adventure | Heist | Action | Indie | Sandbox | Funny | FPS | Realistic | Silent Protagonist | Exploration | RPG | Atmospheric | Immersive Sim | Action | Adventure | Indie | Simulation | Become the best thief. Gather intel, steal things and sell them to buy hi-tech equipment. Do everything that a real thief does. ‖ 사용자 리뷰: 자동차정비 시뮬레이터 생각보다 알차고, 생각보다 부실하다. 할인할때 사면 좋을듯. · 가볍게 즐길만한 감초같은 시뮬레이터 게임 · 할인해서 샀는데 진짜 재미있게 했다. 👍 · 히히 밤에하니 더 꿀잼! 늦잠자겠어요ㅠ 어앉전 해야겠네 · 어 일딴 장점 돈이 잘 벌림 단점 일일이 차 분리 귀찮 노가다!!!!! · 차 분해하는거 시간소요가 꽤 걸리는것 빼고는 전 이 게임 추천합니다. 중독있네요 ㅎㅎ · nice game but when uncle played it robber stole car and crashed and he got caught by police very sad peace 10/10 · Surprisingly Addictive! It's not a Triple-A game, but breaking into houses and stealing toasters is surprisingly fun. The physics can be janky at times, but the progression keeps you hooked. Perfect for when you want to ...</code>	<code>A Way Out | Online Co-Op | Co-op | Split Screen | Story Rich | Local Co-Op | Multiplayer | Action-Adventure | Action | Adventure | Crime | Third Person | Emotional | Dialogue Heavy | Atmospheric | Nudity | Cinematic | Sexual Content | Singleplayer | Violent | Indie | Action | Adventure | Indie | A Way Out is an exclusively co-op adventure where you play the role of one of two prisoners making their daring escape from prison. ‖ 사용자 리뷰: 친구와 둘이 즐길 수 있는 협동 스토리 게임 서로 다른 두 캐릭터가 같은 목표를 갖고 감옥에서 탈출하는 게임인데 차를 타고 다리를 건너갈지, 다리 밑으로 걸어갈지 같은 여러가지 선택지로 플레이 동선이 달라진다 스토리나 캐릭터 상호작용을 보는 맛으로 즐기는 게임이고 초반 탈출 계획을 실행할 때까지 살짝 루즈한 경향이 있다 한국어 지원이 안돼서 싫어하는 사람도 있는데 인터넷에 한국어 패치를 · 재밌구려 슬프구려 반전이 있구려... · 해봤었던 코옵 게임들 중 제일 신선했음 · 친구랑 둘이 하기 참 좋네요 욕 ㅈㄴ먹음 · 마지막 반전이 진짜 충격적이네. 그립습니다 · 조금 억지 엔딩임. 그래도 재미는 있었음. · Don't let the low hour count fool you. I played it with my cousin when It first came out for the EA's launcher. Fantastic co-op game this is! A solid story overall and a ending that will test friendshi...</code>
  <code>TT Isle of Man: Ride on the Edge 2 | Racing | Action | Driving | First-Person | Multiplayer | Simulation | Singleplayer | Automobile Sim | Sports | Realistic | Third Person | Motorbike | Atmospheric | Physics | Co-op | PvP | Family Friendly | RPG | Action | Racing | RPG | Simulation | Sports | Racing across an over 60 km long ultra-technical track at breakneck speed requiring realistic riding skills... that is the challenge that awaits in TT2! Lots of new features lie ahead: open world, reworked physics, classic motorbikes, and more… ‖ 사용자 리뷰: 바이크 게임계의 소울류... TT소울2 어려운데 재미있다. PC로 할꺼면 전체모드 하지말고 윈도우창모드로 하세요. 전체모드 하면 게임엔진의 문제인지 프레임드랍남 · 더트 랠리 바이크 버전. 속도에 심취하면 그대로 트랙 이탈 후 우주로 날아가 버린다. 오토바이에 별 관심이 없어서 그런지 그냥저냥이었다. · 커리어모드 완전 훌륭함 그래픽 봐줄만하고 사운드도 굿 라이드4 바로 삭제해버림 · 2070으로 4k 풀옵 60방어 ㅆㄱㄴ TT 만 라이드 >= motogp22 >>>> ride4 엑박 국밥패드, 조작 난이도 모든 보조 해제 기준 갤 뒤져보고 림스레이싱, 모토지티22, 라이드4, tt 만 2 이렇게 4개 삼. 이거 네 개가 끝인듯? 모토지티랑 라이드는 그래픽뽕에 샀고 tt만 2랑 림스는 좋다해서 삼. 다 해보고 모토지티랑 라이드 빼고 다 환불할라고 했는데 막상해보니까...</code>	<code>Roman Triumph: Survival City Builder | City Builder | Colony Sim | Survival | Crafting | Rome | Open World | Strategy | Singleplayer | Combat | Building | Management | Base Building | Simulation | Sandbox | RTS | 3D | Medieval | Top-Down | God Game | Resource Management | Indie | Simulation | Strategy | At the farthest reaches of the empire, a new Roman city is born! Manage and grow your city, please the gods, and survive against barbarians and mythological threats to achieve Roman Triumph! ‖ 사용자 리뷰: 잘만든게임. 한국어로 제발좀 출시해줬으면... 한국어아니라도 플레이하는데는 별로지장은 없습니다. 재미나요 · 어흑흑 한글 주세요... 현기증 난단 말에요... Pls Korean language · 이거 제발 한국어좀... 재밌을것 같아 · Very promising, one gripe - "dont place houses near quarries as the sound is annoying" Yes, the sound is extremely annoying, so why are you making me listen to it constantly? Please, unless you are zoomed in fairly close · Some similarities to other Roman/ancient civilization citybuilders but with an emphasis on survival. This is Early Access by a...</code>
  <code>The Operational Art of War IV | Strategy | Simulation | Wargame | Turn-Based | World War II | World War I | War | Hex Grid | Cold War | Simulation | Strategy | The Operational Art of War IV is the new generation of operational wargames. With more flexibility than before and with a new array of exciting features, it will make you relive the most iconic battles from the dawn of the 20th Century to modern day, including the ones that never occurred! ‖ 사용자 리뷰: KOR User Guide plz... U can do that at least... ;( · A difficult game that, honestly, I'm not very good at, in spite of the time I've spent on it. I've won at Iwo Jima, but good grief; if you can't beat the Japanese on Iwo Jima, you've got big problems. Still, this is a fu · Thumbs up to make up for the thumbs down from the guy that thought there were only three scenarios. Click the folder button! · Pretty dank. I love nothing more than getting immersed into a deep wargame and sending millions of men to their deaths. My only ...</code>	<code>Nations At War Digital Core Game | Wargame | War | World War II | Military | PvP | Turn-Based | Board Game | Tabletop | Strategy | Top-Down | Simulation | Tactical | Hex Grid | Level Editor | Multiplayer | Singleplayer | Indie | Early Access | Combat | Turn-Based Combat | Indie | Simulation | Strategy | Early Access | Nations At War series is a dynamic platoon-level combat computer game based on the award-winning board games of the same series name. Nations At War series centered on armor, artillery and infantry combat during World War 2. ‖ 사용자 리뷰: I don't usually give reviews with such little play time. But this tactical battles game is a real gem. One of my best purchases of recent times. · The AI is not too smart but that's okay because neither am I. · Just like the Tactical series, this is true to the board games. Plays well, UI is easy to learn and use. Short but good enough tutorial. LNL went more expensive with the DLC's for this series compared to Tactical with · Go...</code>

• Loss: <code>MultipleNegativesRankingLoss</code> with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim",
      "gather_across_devices": false,
      "directions": [
          "query_to_doc"
      ],
      "partition_mode": "joint",
      "hardness_mode": null,
      "hardness_strength": 0.0
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 64
• num_train_epochs: 2
• warmup_steps: 0.1
• bf16: True
• disable_tqdm: True
• dataloader_num_workers: 4

All Hyperparameters

<details><summary>Click to expand</summary>

• per_device_train_batch_size: 64
• num_train_epochs: 2
• max_steps: -1
• learning_rate: 5e-05
• lr_scheduler_type: linear
• lr_scheduler_kwargs: None
• warmup_steps: 0.1
• optim: adamw_torch_fused
• optim_args: None
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• optim_target_modules: None
• gradient_accumulation_steps: 1
• average_tokens_across_devices: True
• max_grad_norm: 1.0
• label_smoothing_factor: 0.0
• bf16: True
• fp16: False
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• use_liger_kernel: False
• liger_kernel_config: None
• use_cache: False
• neftune_noise_alpha: None
• torch_empty_cache_steps: None
• auto_find_batch_size: False
• log_on_each_node: True
• logging_nan_inf_filter: True
• include_num_input_tokens_seen: no
• log_level: passive
• log_level_replica: warning
• disable_tqdm: True
• project: huggingface
• trackio_space_id: None
• trackio_bucket_id: None
• trackio_static_space_id: None
• per_device_eval_batch_size: 8
• prediction_loss_only: True
• eval_on_start: False
• eval_do_concat_batches: True
• eval_use_gather_object: False
• eval_accumulation_steps: None
• include_for_metrics: []
• batch_eval_metrics: False
• save_only_model: False
• save_on_each_node: False
• enable_jit_checkpoint: False
• push_to_hub: False
• hub_private_repo: None
• hub_model_id: None
• hub_strategy: every_save
• hub_always_push: False
• hub_revision: None
• load_best_model_at_end: False
• ignore_data_skip: False
• restore_callback_states_from_checkpoint: False
• full_determinism: False
• seed: 42
• data_seed: None
• use_cpu: False
• accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
• parallelism_config: None
• dataloader_drop_last: False
• dataloader_num_workers: 4
• dataloader_pin_memory: True
• dataloader_persistent_workers: False
• dataloader_prefetch_factor: None
• remove_unused_columns: True
• label_names: None
• train_sampling_strategy: random
• length_column_name: length
• ddp_find_unused_parameters: None
• ddp_bucket_cap_mb: None
• ddp_broadcast_buffers: False
• ddp_static_graph: None
• ddp_backend: None
• ddp_timeout: 1800
• fsdp: []
• fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
• deepspeed: None
• debug: []
• skip_memory_metrics: True
• do_predict: False
• resume_from_checkpoint: None
• warmup_ratio: None
• local_rank: -1
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

</details>

Training Logs

<details><summary>Click to expand</summary> </details>

Training Time

• Training: 3.9 hours

Framework Versions

• Python: 3.12.3
• Sentence Transformers: 5.6.0
• Transformers: 5.7.0
• PyTorch: 2.11.0+cu128
• Accelerate: 1.13.0
• Datasets: 5.0.0
• Tokenizers: 0.22.2

Citation

BibTeX

Sentence Transformers

@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",
}

MultipleNegativesRankingLoss

@misc{oord2019representationlearningcontrastivepredictive,
      title={Representation Learning with Contrastive Predictive Coding},
      author={Aaron van den Oord and Yazhe Li and Oriol Vinyals},
      year={2019},
      eprint={1807.03748},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/1807.03748},
}

<!-- ## 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.* -->