Upload README.md with huggingface_hub

README.md

@@ -1,199 +1,227 @@
 ---
+license: apache-2.0
+base_model:
+- swiss-ai/Apertus-70B-2509
+pipeline_tag: text-generation
 library_name: transformers
-tags: []
+tags:
+- multilingual
+- compliant
+- swiss-ai
+- apertus
+- heretic
+- uncensored
+- decensored
+- abliterated
+extra_gated_prompt: "### Apertus LLM Acceptable Use Policy  \n(1.0 | September 1,\
+  \ 2025)\n\"Agreement\" The Swiss National AI Institute (SNAI) is a partnership between\
+  \ the two Swiss Federal Institutes of Technology, ETH Zurich and EPFL. \n\nBy using\
+  \ the Apertus LLM you agree to indemnify, defend, and hold harmless ETH Zurich and\
+  \ EPFL against any third-party claims arising from your use of Apertus LLM. \n\n\
+  The training data and the Apertus LLM may contain or generate information that directly\
+  \ or indirectly refers to an identifiable individual (Personal Data). You process\
+  \ Personal Data as independent controller in accordance with applicable data protection\
+  \ law. SNAI will regularly provide a file with hash values for download which you\
+  \ can apply as an output filter to your use of our Apertus LLM. The file reflects\
+  \ data protection deletion requests which have been addressed to SNAI as the developer\
+  \ of the Apertus LLM. It allows you to remove Personal Data contained in the model\
+  \ output. We strongly advise downloading and applying this output filter from SNAI\
+  \ every six months following the release of the model.  "
+extra_gated_fields:
+  Your Name: text
+  Country: country
+  Affiliation: text
+  geo: ip_location
+  By clicking Submit below I accept the terms of use: checkbox
+extra_gated_button_content: Submit
 ---
+# This is a decensored version of [swiss-ai/Apertus-70B-Instruct-2509](https://huggingface.co/swiss-ai/Apertus-70B-Instruct-2509), made using [Heretic](https://github.com/p-e-w/heretic) v1.2.0
 
-# Model Card for Model ID
+## Abliteration parameters
 
-<!-- Provide a quick summary of what the model is/does. -->
+| Parameter | Value |
+| :-------- | :---: |
+| **direction_index** | per layer |
+| **attn.o_proj.max_weight** | 0.83 |
+| **attn.o_proj.max_weight_position** | 50.26 |
+| **attn.o_proj.min_weight** | 0.12 |
+| **attn.o_proj.min_weight_distance** | 39.50 |
+| **mlp.down_proj.max_weight** | 0.92 |
+| **mlp.down_proj.max_weight_position** | 75.69 |
+| **mlp.down_proj.min_weight** | 0.38 |
+| **mlp.down_proj.min_weight_distance** | 20.28 |
 
+## Performance
 
+| Metric | This model | Original model ([swiss-ai/Apertus-70B-Instruct-2509](https://huggingface.co/swiss-ai/Apertus-70B-Instruct-2509)) |
+| :----- | :--------: | :---------------------------: |
+| **KL divergence** | 0.1916 | 0 *(by definition)* |
+| **Refusals** | 25/100 | 99/100 |
 
-## Model Details
+-----
 
-### Model Description
 
-<!-- Provide a longer summary of what this model is. -->
+# Apertus
 
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
+![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/6639f08490b7db8dcbf1a2aa/YKux3SpTciL4O60L3Ol-6.jpeg)
 
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
+##  Table of Contents
 
-### Model Sources [optional]
+1. [Model Summary](#model-summary)
+2. [How to use](#how-to-use)
+3. [Evaluation](#evaluation)
+4. [Training](#training)
+5. [Limitations](#limitations)
+6. [Legal Aspects](#legal-aspects)
 
-<!-- Provide the basic links for the model. -->
+## Model Summary
 
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
+Apertus is a 70B and 8B parameter language model designed to push the boundaries of fully-open multilingual and transparent models. 
+The model supports over 1000 languages and long context, it uses only fully compliant and open training data, and achieves comparable performance to models trained behind closed doors.
 
-## Uses
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/654baf61d625e083383dfd00/gKDv_6dpIpvmgyquenbXt.png)
 
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+The model is a decoder-only transformer, pretrained on 15T tokens with a staged curriculum of web, code and math data. The model uses a new xIELU activation function and is trained from scratch with the AdEMAMix optimizer. Post-training included supervised fine-tuning and alignment via QRPO.
 
-### Direct Use
+### Key features
+- **Fully open model**: open weights + open data + full training details including all data and training recipes
+- **Massively Multilingual**: 1811 natively supported languages
+- **Compliant** Apertus is trained while respecting opt-out consent of data owners (even retrospectivey), and avoiding memorization of training data
 
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+For more details refer to our [technical report](https://arxiv.org/abs/2509.14233)
 
-[More Information Needed]
+## How to use
 
-### Downstream Use [optional]
+The modeling code for Apertus is available in transformers `v4.56.0` and later, so make sure to upgrade your transformers version. You can also load the model with the latest `vLLM` which uses transformers as a backend.
+```bash
+pip install -U transformers
+```
 
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-[More Information Needed]
+model_name = "swiss-ai/Apertus-70B-Instruct-2509"
+device = "cuda"  # for GPU usage or "cpu" for CPU usage
 
-### Out-of-Scope Use
+# load the tokenizer and the model
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+).to(device)
 
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+# prepare the model input
+prompt = "Give me a brief explanation of gravity in simple terms."
+messages_think = [
+    {"role": "user", "content": prompt}
+]
 
-[More Information Needed]
+text = tokenizer.apply_chat_template(
+    messages_think,
+    tokenize=False,
+    add_generation_prompt=True,
+)
+model_inputs = tokenizer([text], return_tensors="pt", add_special_tokens=False).to(model.device)
 
-## Bias, Risks, and Limitations
+# Generate the output
+generated_ids = model.generate(**model_inputs, max_new_tokens=32768)
 
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+# Get and decode the output
+output_ids = generated_ids[0][len(model_inputs.input_ids[0]) :]
+print(tokenizer.decode(output_ids, skip_special_tokens=True))
+```
 
-[More Information Needed]
+>[!TIP]
+> We recommend setting `temperature=0.8` and `top_p=0.9` in the sampling parameters.
 
-### Recommendations
+### Long context processing
 
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+Apertus by default supports a context length up to 65,536 tokens.
 
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+### Agentic Usage
 
-## How to Get Started with the Model
+Apertus supports tool use
 
-Use the code below to get started with the model.
+### Deployment
 
-[More Information Needed]
-
-## Training Details
-
-### Training Data
-
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-
-[More Information Needed]
-
-### Training Procedure
-
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-
-#### Preprocessing [optional]
-
-[More Information Needed]
-
-
-#### Training Hyperparameters
-
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-
-#### Speeds, Sizes, Times [optional]
-
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-
-[More Information Needed]
+Deployment of the models is directly supported by the newest versions of [Transformers](https://github.com/huggingface/transformers), [vLLM](https://github.com/vllm-project/vllm), [SGLang](https://github.com/sgl-project/sglang), and also for running on-device with [MLX](https://github.com/ml-explore/mlx-lm), 
 
 ## Evaluation
 
-<!-- This section describes the evaluation protocols and provides the results. -->
-
-### Testing Data, Factors & Metrics
-
-#### Testing Data
-
-<!-- This should link to a Dataset Card if possible. -->
-
-[More Information Needed]
-
-#### Factors
-
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
-
-[More Information Needed]
-
-#### Metrics
-
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-
-[More Information Needed]
-
-### Results
-
-[More Information Needed]
-
-#### Summary
-
-
-
-## Model Examination [optional]
-
-<!-- Relevant interpretability work for the model goes here -->
-
-[More Information Needed]
-
-## Environmental Impact
-
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
-
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
-
-## Technical Specifications [optional]
-
-### Model Architecture and Objective
-
-[More Information Needed]
-
-### Compute Infrastructure
-
-[More Information Needed]
-
-#### Hardware
-
-[More Information Needed]
-
-#### Software
-
-[More Information Needed]
-
-## Citation [optional]
-
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
-
-**BibTeX:**
-
-[More Information Needed]
-
-**APA:**
-
-[More Information Needed]
-
-## Glossary [optional]
-
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
-
-[More Information Needed]
-
-## More Information [optional]
-
-[More Information Needed]
-
-## Model Card Authors [optional]
-
-[More Information Needed]
-
-## Model Card Contact
-
-[More Information Needed]
+**Pretraining Evaluation:** Performance (%) of Apertus models on *general language understanding* tasks (higher is better) compared to other pretrained models.
+
+| **Model** | **Avg** | **ARC** | **HellaSwag** | **WinoGrande** | **XNLI** | **XCOPA** | **PIQA** |
+| :--- | :---: | :---: | :---: | :---: | :---: | :---: | :---: |
+| **Fully Open Models** | | | | | | | |
+| **Apertus-8B** | 65.8 | 72.7 | 59.8 | 70.6 | 45.2 | 66.5 | 79.8 |
+| **Apertus-70B** | 67.5 | 70.6 | 64.0 | 73.3 | 45.3 | 69.8 | 81.9 |
+| OLMo2-7B | 64.0 | 72.9 | 60.4 | 74.5 | 40.4 | 55.2 | 80.9 |
+| OLMo2-32B | 67.7 | 76.2 | 66.7 | 78.6 | 42.9 | 60.1 | 82.1 |
+| EuroLLM-1.7B | 54.8 | 57.2 | 44.9 | 58.1 | 40.7 | 55.7 | 72.4 |
+| EuroLLM-9B | 62.8 | 67.9 | 57.9 | 68.8 | 41.5 | 61.1 | 79.6 |
+| SmolLM2-1.7B | 58.5 | 66.1 | 52.4 | 65.6 | 37.6 | 52.3 | 77.0 |
+| SmolLM3-3B | 61.6 | 68.6 | 56.4 | 68.1 | 40.5 | 58.2 | 77.7 |
+| Poro-34B | 61.7 | 65.7 | 57.9 | 70.6 | 41.6 | 56.0 | 78.5 |
+| **Open-Weight Models** | | | | | | | |
+| Llama3.1-8B | 65.4 | 71.6 | 60.0 | 73.4 | 45.3 | 61.8 | 80.1 |
+| Llama3.1-70B | 67.3 | 74.4 | 56.5 | 79.4 | 44.3 | 66.7 | 82.3 |
+| Qwen2.5-7B | 64.4 | 69.6 | 60.1 | 72.8 | 43.3 | 61.7 | 78.7 |
+| Qwen2.5-72B | 69.8 | 76.2 | 67.5 | 78.0 | 46.9 | 68.2 | 82.0 |
+| Qwen3-32B | 67.8 | 75.6 | 64.0 | 73.8 | 44.4 | 67.9 | 80.9 |
+| Llama4-Scout-16x17B | 67.9 | 74.7 | 66.8 | 73.2 | 43.5 | 67.7 | 81.2 |
+| GPT-OSS-20B | 58.1 | 67.0 | 41.5 | 66.5 | 37.4 | 60.4 | 75.6 |
+
+Many additional benchmark evaluations, for pretraining and posttraining phases, multilingual evaluations in around hundred languages, and long context evaluations are provided in Section 5 of the [Apertus_Tech_Report.pdf](https://github.com/swiss-ai/apertus-tech-report/blob/main/Apertus_Tech_Report.pdf)
+
+## Training
+
+### Model
+
+- **Architecture:** Transformer decoder
+- **Pretraining tokens:** 15T
+- **Precision:** bfloat16
+
+### Software & hardware
+
+- **GPUs:** 4096 GH200
+- **Training Framework:** [Megatron-LM](https://github.com/swiss-ai/Megatron-LM)
+- ...
+
+### Open resources
+All elements used in the training process are made openly available
+- **Training data reconstruction scripts:** [github.com/swiss-ai/pretrain-data](https://github.com/swiss-ai/pretrain-data)
+- The training intermediate checkpoints are available on the different branches of this same repository
+
+
+## Limitations
+
+Apertus can produce text on a variety of topics, but the generated content may not always be factually accurate, logically consistent, or free from biases present in the training data. These models should be used as assistive tools rather than definitive sources of information. Users should always verify important information and critically evaluate any generated content.
+
+
+## Legal Aspects
+
+#### EU AI Act Transparency Documentation and Code of Practice
+- [Apertus_EU_Public_Summary.pdf](https://huggingface.co/swiss-ai/Apertus-70B-2509/blob/main/Apertus_EU_Public_Summary.pdf)
+- [Apertus_EU_Code_of_Practice.pdf](https://huggingface.co/swiss-ai/Apertus-70B-2509/blob/main/Apertus_EU_Code_of_Practice.pdf)
+
+#### Data Protection and Copyright Requests
+For removal requests of personally identifiable information (PII) or of copyrighted content, please contact the respective dataset owners or us directly
+- llm-privacy-requests@swiss-ai.org
+- llm-copyright-requests@swiss-ai.org
+  
+#### Output Filter for PII
+- Currently no output filter is provided.
+- Please check this site regularly for an output filter that can be used on top of the Apertus LLM. The filter reflects data protection deletion requests which have been addressed to us as the developer of the Apertus LLM. It allows you to remove Personal Data contained in the model output. We strongly advise downloading and applying this output filter from this site every six months.
+
+## Contact
+To contact us, please send an email to
+llm-requests@swiss-ai.org
+
+## Citation
+```bash
+@misc{swissai2025apertus,
+  title={{Apertus: Democratizing Open and Compliant LLMs for Global Language Environments}},
+  author={Alejandro Hernández-Cano and Alexander Hägele and Allen Hao Huang and Angelika Romanou and Antoni-Joan Solergibert and Barna Pasztor and Bettina Messmer and Dhia Garbaya and Eduard Frank Ďurech and Ido Hakimi and Juan García Giraldo and Mete Ismayilzada and Negar Foroutan and Skander Moalla and Tiancheng Chen and Vinko Sabolčec and Yixuan Xu and Michael Aerni and Badr AlKhamissi and Ines Altemir Marinas and Mohammad Hossein Amani and Matin Ansaripour and Ilia Badanin and Harold Benoit and Emanuela Boros and Nicholas Browning and Fabian Bösch and Maximilian Böther and Niklas Canova and Camille Challier and Clement Charmillot and Jonathan Coles and Jan Deriu and Arnout Devos and Lukas Drescher and Daniil Dzenhaliou and Maud Ehrmann and Dongyang Fan and Simin Fan and Silin Gao and Miguel Gila and María Grandury and Diba Hashemi and Alexander Hoyle and Jiaming Jiang and Mark Klein and Andrei Kucharavy and Anastasiia Kucherenko and Frederike Lübeck and Roman Machacek and Theofilos Manitaras and Andreas Marfurt and Kyle Matoba and Simon Matrenok and Henrique Mendoncça and Fawzi Roberto Mohamed and Syrielle Montariol and Luca Mouchel and Sven Najem-Meyer and Jingwei Ni and Gennaro Oliva and Matteo Pagliardini and Elia Palme and Andrei Panferov and Léo Paoletti and Marco Passerini and Ivan Pavlov and Auguste Poiroux and Kaustubh Ponkshe and Nathan Ranchin and Javi Rando and Mathieu Sauser and Jakhongir Saydaliev and Muhammad Ali Sayfiddinov and Marian Schneider and Stefano Schuppli and Marco Scialanga and Andrei Semenov and Kumar Shridhar and Raghav Singhal and Anna Sotnikova and Alexander Sternfeld and Ayush Kumar Tarun and Paul Teiletche and Jannis Vamvas and Xiaozhe Yao and Hao Zhao Alexander Ilic and Ana Klimovic and Andreas Krause and Caglar Gulcehre and David Rosenthal and Elliott Ash and Florian Tramèr and Joost VandeVondele and Livio Veraldi and Martin Rajman and Thomas Schulthess and Torsten Hoefler and Antoine Bosselut and Martin Jaggi and Imanol Schlag},
+  year={2025},
+  howpublished={\url{https://arxiv.org/abs/2509.14233}}
+}
+```