Add SetFit model

1_Pooling/config.json

@@ -0,0 +1,10 @@
+{
+  "word_embedding_dimension": 768,
+  "pooling_mode_cls_token": false,
+  "pooling_mode_mean_tokens": true,
+  "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 +1,288 @@
+---
+language: en
+license: apache-2.0
+library_name: setfit
+tags:
+- setfit
+- sentence-transformers
+- text-classification
+- generated_from_setfit_trainer
+base_model: sentence-transformers/paraphrase-mpnet-base-v2
+metrics:
+- accuracy
+- precision
+- recall
+- f1
+widget:
+- text: a cinematic masterpiece.. i can't believe my eyes what just i experienced
+    in theatre, this movie is one of the best cinematic experience i've ever had..
+    from opening scene to end credit every second is worth it.. epic score by hans
+    zimmer elevates the experience, filled with beautiful shots and cinematography
+    to a great scrrenplay and some top notch acting by every cast member dune 2 will
+    be remembered as one of the best films of this generation even after 50 years
+    later that's my guarantee.. denis villeneuve is one of the best director of this
+    generation who gave us masterpiece like blade runner 2049 and currently working
+    on 3rd part of this trilogy which i can't wait to see.. action scenes were great
+    featuring big ships, blasters laser guns and every other kind of weapons, sound
+    effects were amazing but most importantly screenplay was well paced and story
+    featuring blend of sci fi and fantasy elements is absolute best of the best even
+    though it is an adaptation of 1965 novel of same name it is one of the best adaptations
+    ever.. i loved both movies and can't wait to see what happens in future because
+    i enjoy films about the chosen one prophecy like star wars, the matrix and now
+    this.. i won't give any spoiler here and will not talk about anything shown in
+    the movie since you need to see this movie for yourself to feel the magic.. don't
+    miss it at any cost.. watch it on the biggest screen possible especially imax
+    to enjoy epic sound effects and see the larger than life picture you'll love it..
+- text: hopeful for the future of this franchise this one's very personal for me since
+    i've been invested in this franchise for years now and all the waiting throughout
+    the production of both parts, considering the multiple delays, have been worthwhile
+    and it only enriched the experience, so it's really hard for me to not sound like
+    an absolute fanboy.every tidbit of news from the start of the production only
+    fueled my excitement, from cast and crew announcements to first look reveals to
+    teasers and trailers which i probably watched and obsessively dissected dozens
+    of times. which always put at rest my lore obsessed brain because it all made
+    sense, i could tell from the beginning that both parts were crafted by people
+    who understood and cared about the source material without approaching it like
+    some sort of sacred text. on that note i was quite surprised and pleased by the
+    changes they did with characters, timeline and most specifically the ending, which
+    only left me wanting more.despite the fact that unfortunately my local cinema
+    doesn't have imax screens the experience was an absolute blast and i was nothing
+    short of euphoric with a stupid grin the entire movie it felt surreal and overwhelming
+    at times taking it all in. i loved how it expanded from the themes established
+    in the first one, particularly the detailed exploration of fremen culture, how
+    ominous and mysterious felt the bene gesserit, as for the harkonnens besides the
+    obvious praise for how brutal, weird and cool they're portrayed i particularly
+    appreciated the h. r giger inspired look of their black sun lit giedi prime. all
+    while staying true to the essence of the story.with a chunky and lengthy runtime
+    its impossible for me to pinpoint favorite moments. my jaw dropped at the sheer
+    scale of many sequences, namely the opening scene with the orange hues and the
+    levitating harkonnens, also the gladiator scene, the big feyd-rautha entrance
+    was bonkers the fighting choreography felt natural and realistic and well thought
+    out, tho i gotta admit i was iffy at first but ultimately pleasantly surprised
+    by austin's feyd-rautha.overall, the performances, direction, music, sound design,
+    cinematography, and production design were outstanding. the themes of fanaticism
+    and the difficulty of reasoning with true believers resonated deeply with me.
+    plus, getting to watch it with three of my favorite people made the movie watching
+    experience even more lovely. dune part two exceeded my every expectations and
+    left me eagerly awaiting the next installment. i can confidently say that this
+    is the perfect dune adaptation.
+- text: disappointing like seriously... what the hell was that? :-( first part was
+    mostly on point... but the second part?? why alter perfection?it totally missed
+    in my opinion. the nukes were used for the shields, not for the spice. the water
+    of life was meant to be used for the spice... the great houses came together with
+    the emperor, why was this changed? where is the baby? where is the sister? why
+    was the guild cut out all together? where is the scene between. gurney and lady
+    jessica. all this gave context to the story what now went totally missing.as it
+    was cut together now, the only thing truly great about the movie was the sound
+    editing. this gave a great atmosphere. i in general like villeneuve alot, sicario
+    is one of the top3 the greatest movies ever made, but this was disappointing.
+- text: dune 1 & 2 are probably the best sci-fi movies ever... yes, dune 1 & 2 are
+    probably the best sci-fi movies ever. i can't really come up with anything else
+    that would beat these anyway.part two was a nice continuation to part one, however
+    i felt the last part of the movie could have been better or maybe it just went
+    too fast. without spoiling its hard to explain. also missed some more insight
+    about the mother and a lot of other things as well... imagine if this could have
+    been a 20 episode serie with same quality!anyhow visually and the main cast were
+    just excellent. i did expect more out of the emperor christofer walken.just love
+    the tonality of the entire movie and the sound and music made it masterful.please
+    more of these "adult" epic sci-fi movies and series!
+- text: great movie with holes in the plot dune 2 is a great movie that can be enjoyed
+    despite some holes in the plot and questionable story elements. the movie is great
+    visually and serves as an example of how cgi doesn't have to make a movie seem
+    sterile and fake, or like watching a video game like the new brand of marvel movies
+    and aquaman do.the cast is great - christopher walken should have had more scenes
+    and other than it being him, his performance was not unique in any way - and there
+    is a non credited appearance that i'm not going to spoil that is a wow!the movie
+    is long, but there's a difference between long and too long, and any movie that
+    wants to be an epic. 2 out of 5 found this helpful. was this review helpful? sign
+    in to vote. permalink
+pipeline_tag: text-classification
+inference: true
+model-index:
+- name: SetFit with sentence-transformers/paraphrase-mpnet-base-v2 on data/raw/15239678.jsonl
+  results:
+  - task:
+      type: text-classification
+      name: Text Classification
+    dataset:
+      name: data/raw/15239678.jsonl
+      type: unknown
+      split: test
+    metrics:
+    - type: accuracy
+      value: 0.8246753246753247
+      name: Accuracy
+    - type: precision
+      value: 0.9915611814345991
+      name: Precision
+    - type: recall
+      value: 0.818815331010453
+      name: Recall
+    - type: f1
+      value: 0.8969465648854962
+      name: F1
+---
+
+# SetFit with sentence-transformers/paraphrase-mpnet-base-v2 on data/raw/15239678.jsonl
+
+This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [sentence-transformers/paraphrase-mpnet-base-v2](https://huggingface.co/sentence-transformers/paraphrase-mpnet-base-v2) as the Sentence Transformer embedding model. A [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance is used for classification.
+
+The model has been trained using an efficient few-shot learning technique that involves:
+
+1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
+2. Training a classification head with features from the fine-tuned Sentence Transformer.
+
+## Model Details
+
+### Model Description
+- **Model Type:** SetFit
+- **Sentence Transformer body:** [sentence-transformers/paraphrase-mpnet-base-v2](https://huggingface.co/sentence-transformers/paraphrase-mpnet-base-v2)
+- **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
+- **Maximum Sequence Length:** 512 tokens
+- **Number of Classes:** 2 classes
+<!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
+- **Language:** en
+- **License:** apache-2.0
+
+### Model Sources
+
+- **Repository:** [SetFit on GitHub](https://github.com/huggingface/setfit)
+- **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055)
+- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
+
+### Model Labels
+| Label    | Examples                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               |
+|:---------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+| negative | <ul><li>"it's like watching a cult and not really a good movie. sorry not for me. all the characters gave justice on their roles, well executed. the pace of the story is so slow that it made me so sleepy in the middle of the film. actually the film made me bored coz i expected more but a lousy story. i would like to mention rebecca ferguson, she's a great actress and really fit to be the lady jessica itself. in the middle of the film, me and three of my friends wanted to go home coz it feels like we're watching a cult movie. not a good one for children's to watch. in this movie, nothing much to expect, such a waste of time and it was a very boring film fighting for a spice. 19 out of 100 found this helpful. was this review helpful? sign in to vote. permalink"</li><li>'the avengers-era "cinema" for a call of duty generation i watched dune: part two on imax (as seems befitting that sort of big-budget picture), in pretty good seats considering the house was packed with a diverse crowd of young adults for the late-night showing.my review of the first of denis villeneuve\'s adaptations was that it was pretty to look at, but basically uninteresting in its plot, a major let-down considering how hyped the books have been, when on the screen they\'re no better than star wars. this second one exhibits those same features, but worse.this is basically an action movie--a genre which i\'m fundamentally uninterested in--which seems at odds with all the talk of dune being characterized by complex political & economic maneuvering--of which this film does include some, albeit parenthetically. the scenes are impressionistic in their pacing and content, and expressionistic in that the photography & score are more about attempting to convey emotionality, or at least get the adrenaline pumping, than any sort of profound characterizations or themes. this is lawrence of arabia in space but with the cold war as the contemporary allegory. that sort of story would still have been resonant during the post-9/11 era, but we\'ve moved on from those wars by now into an altogether different geopolitics and political economy.the monumentalism of the shots can be impressive, and certainly the design of the mechanical vehicles is (in addition to the still-eye candy costumes), but there are too many moments where it\'s like watching a video game cut-scene, or even a video game playing out (c.f. the entire opening sequence). cinema is going backwards when that\'s the case. and there are some egregious dissonances in the sound design, most of all in the various accents the actors were allowed to use, which takes us out of the suspension of disbelief, but also in the over-application of non-diegetic score, which treats the audience as dumb emotions-receptacles, during scenes that by-and-large do not earn such responses.(the only outstanding sequence of the film is when paul attempts his first worm-ride, which is almost as exhilirating as the roller coaster ride it attempts to be for us on screen.)there are also weird dissonances in this film attempting to be humorous, as the on-again off-again nature of those lines (and how they\'re delivered) feel like a wink to the audience that this allegory of abrahamic religions co-opting a people\'s revolution--albeit against a totalitarian capitalist empire, mixing all kinds of economic stages of history, much as the genres of swords & planets and sci-fi and fantasy are mixed--is not to be taken too seriously in our secular times. but then, what are we to take from this "epic"?'</li><li>"try as hard as you like i left feeling bored, and that's a bad sign for an epic sci-fi. i watched dune i and loved it, the deep sonorous sounds of an alien nation were exciting, and the dialogue kept minimal and thick. dune ii blasts you with ott noise, over the top effects, and piss poor dialogue... 'for the duke, and my friends' i felt this movie was aimed at a pre-teen audience with spiderman fetish and a heavy hand of americanism... the emperor was not an emperor. recommend putting a human written storyline first and leave ai/cgi to a bare minimum. i watched dune 20 years ago and loved it. i read dune 30 years ago and loved it. the actors tried hard with what they got but it wasn't enough to make it a masterpiece, it's more like an aprenticepiece."</li></ul> |
+| positive | <ul><li>"best movie in recent years. the movie is a more impressive, more exciting and deeper production than the first part. give it 9 out of 10.the acting, visual effects, music and atmosphere in the movie are amazing. paul's evolution is perfectly demonstrated by timothee chalamet.the visual effects deliver a breathtaking rendering of arrakis' desert landscapes, giant worms, and battle scenes. you can smell the scent of a spice behind curtain.music adds soul to the film with the mastery of hans zimmer. the atmosphere is to keep tension and emotion at a high level throughout the film.while the film's script remains faithful to the book, it also offers some differences. while some changes make the film better, others do not fit the spirit of the book. however, i can honestly say that the changes made do not harm the film at all.dune: part two is one of the masterpieces of science fiction cinema. it's the best movie i've seen in nearly 15 years. maybe oppenheimer.nah, this movie is better."</li><li>"best movie ever what a movie!!!!!!!!!!the whole movie is so good! the setting is amazing. the story is so good. and the plot and the premise is amazing my favorite characters are paul and chani. this whole movie is seeking revenge from the darkness and the evil that was causing the world to collapse. and following that they decide just like on star wars to choose what side they're on the light side or the dark side. and man the cliffhanger of the movie was the most shocking thing ever. the whole movie is a very well done movie. i love this movie so much it gave me chills. this movie will get the best picture next year for the oscars. 0 out of 0 found this helpful. was this review helpful? sign in to vote. permalink"</li><li>"very spicey! undoubtedly one of the greatest cinematic experiences i've ever had. most movies nowadays try to wow you with either visuals, action, character development, and seemingly less and less so everyday with a reasonable plot. this movie manages to check off every box. it is hard to pinpoint an actor in it that wasn't superb in their role. the visual effects, especially in imax , are stunning and are not simply artistic in nature, but actually work to advance the plot.the film is pieced together so brilliantly with each act seamlessly fitting in with the last. the score by zimmer is unsurprisingly absolute perfection as well.for the length of the movie it actually feels quite short which is a huge achievement and shows how immersed you become in villneuve's world.however, i would have liked to see more of feyd rautha. they had an epic scene of his in the harkonnen arena, but not too much else to develop him for his final battle against paul. i also would have liked to see the sardaukar put up a little better of a fight. they somehow went from the empires greatest fighters to pretty pathetic. perhaps this was to show how strong paul and his forces had become, but it just seemed a bit too easy. still these are just personal aspects i would have liked to see and are not issues in the movie itself.overall, an epic movie, and the trilogy is likely to be sitting right alongside the lord of the rings and the original star wars movies when all is said and done.this is a can't miss in imax."</li></ul>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                             |
+
+## Evaluation
+
+### Metrics
+| Label   | Accuracy | Precision | Recall | F1     |
+|:--------|:---------|:----------|:-------|:-------|
+| **all** | 0.8247   | 0.9916    | 0.8188 | 0.8969 |
+
+## Uses
+
+### Direct Use for Inference
+
+First install the SetFit library:
+
+```bash
+pip install setfit
+```
+
+Then you can load this model and run inference.
+
+```python
+from setfit import SetFitModel
+
+# Download from the 🤗 Hub
+model = SetFitModel.from_pretrained("carlesoctav/SentimentClassifierDune-8shot")
+# Run inference
+preds = model("great movie with holes in the plot dune 2 is a great movie that can be enjoyed despite some holes in the plot and questionable story elements. the movie is great visually and serves as an example of how cgi doesn't have to make a movie seem sterile and fake, or like watching a video game like the new brand of marvel movies and aquaman do.the cast is great - christopher walken should have had more scenes and other than it being him, his performance was not unique in any way - and there is a non credited appearance that i'm not going to spoil that is a wow!the movie is long, but there's a difference between long and too long, and any movie that wants to be an epic. 2 out of 5 found this helpful. was this review helpful? sign in to vote. permalink")
+```
+
+<!--
+### Downstream Use
+
+*List how someone could finetune this model on their own dataset.*
+-->
+
+<!--
+### 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 Set Metrics
+| Training set | Min | Median | Max |
+|:-------------|:----|:-------|:----|
+| Word count   | 121 | 271.5  | 993 |
+
+| Label    | Training Sample Count |
+|:---------|:----------------------|
+| negative | 8                     |
+| positive | 8                     |
+
+### Training Hyperparameters
+- batch_size: (16, 16)
+- num_epochs: (1, 1)
+- max_steps: -1
+- sampling_strategy: oversampling
+- body_learning_rate: (2e-05, 1e-05)
+- head_learning_rate: 0.01
+- loss: CosineSimilarityLoss
+- distance_metric: cosine_distance
+- margin: 0.25
+- end_to_end: False
+- use_amp: False
+- warmup_proportion: 0.1
+- seed: 42
+- eval_max_steps: -1
+- load_best_model_at_end: True
+
+### Training Results
+| Epoch   | Step  | Training Loss | Validation Loss |
+|:-------:|:-----:|:-------------:|:---------------:|
+| 0.1111  | 1     | 0.2058        | -               |
+| **1.0** | **9** | **-**         | **0.2368**      |
+
+* The bold row denotes the saved checkpoint.
+### Framework Versions
+- Python: 3.10.11
+- SetFit: 1.0.3
+- Sentence Transformers: 2.5.1
+- Transformers: 4.38.2
+- PyTorch: 2.0.1
+- Datasets: 2.18.0
+- Tokenizers: 0.15.2
+
+## Citation
+
+### BibTeX
+```bibtex
+@article{https://doi.org/10.48550/arxiv.2209.11055,
+    doi = {10.48550/ARXIV.2209.11055},
+    url = {https://arxiv.org/abs/2209.11055},
+    author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
+    keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
+    title = {Efficient Few-Shot Learning Without Prompts},
+    publisher = {arXiv},
+    year = {2022},
+    copyright = {Creative Commons Attribution 4.0 International}
+}
+```
+
+<!--
+## 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 @@
+{
+  "_name_or_path": "exp/dune2-sent-classfier-8-shot/step_9",
+  "architectures": [
+    "MPNetModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "mpnet",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "relative_attention_num_buckets": 32,
+  "torch_dtype": "float32",
+  "transformers_version": "4.38.2",
+  "vocab_size": 30527
+}

config_sentence_transformers.json

@@ -0,0 +1,9 @@
+{
+  "__version__": {
+    "sentence_transformers": "2.0.0",
+    "transformers": "4.7.0",
+    "pytorch": "1.9.0+cu102"
+  },
+  "prompts": {},
+  "default_prompt_name": null
+}

config_setfit.json

@@ -0,0 +1,7 @@
+{
+  "normalize_embeddings": false,
+  "labels": [
+    "negative",
+    "positive"
+  ]
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b23d44a901b90d56e7f1c2f67e3d2082dede6e4c7fe8b4c42e080b978b65b80
+size 437967672

model_head.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ab948312a0da79000ae118a0f9ef0933b96a21d5412aa2814d43f7688bb0700
+size 6997

modules.json

@@ -0,0 +1,14 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  }
+]

sentence_bert_config.json

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

special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "</s>",
+    "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 +1,66 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "104": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "30526": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "max_length": 512,
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_to_multiple_of": null,
+  "pad_token": "<pad>",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "sep_token": "</s>",
+  "stride": 0,
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "MPNetTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "[UNK]"
+}