readme update

README.md

@@ -1,202 +1,84 @@
 ---
 base_model: unsloth/llama-3.2-1b-instruct-bnb-4bit
 library_name: peft
+language:
+  - es
+  - en
+tags:
+  - llama-3
+  - llama
+  - meta
+  - facebook
+  - unsloth
+  - deepsphere
+  - deepsphereAI
+  - transformers
 ---
 
 # Model Card for Model ID
-
-<!-- Provide a quick summary of what the model is/does. -->
-
-
+El Modelo RaMem (Rag as Memory) implementa el enfoque de usar RAG para potenciar modelos pequeños y hacerlos muy competentes para el hardware limitado para el que son diseñados.
 
 ## Model Details
 
 ### Model Description
+RaMen es un Modelo de Lenguaje Pequeño, con la particularidad de que es muy poderoso, ya que integra Rag como Memoria (RaMen). Este enfoque proviene de la limitación de la ventana de contexto de los modelos de lenguaje, usando un sistema RAG para recuperar información relevante y añadirla como contexto esto se soluciona en parte.
 
-<!-- Provide a longer summary of what this model is. -->
+El modelo fue fine-tuneado con una combinación de datasets, por una parte está el OASST_2 dataset y por otra parte un dataset generado sinteticamente con ayuda de modelos como DeepSeek y ChatGPT logrando así crear un gran dataset, la finalidad de este dataset no es que el modelo aprenda a generar buenas respuestas, ya que para eso está el enfoque RAG, sino que el modelo pueda afinar su conocimiento sobre la lengua española para evitar lo más posible incoherencias en la respuesta.
 
 
-
-- **Developed by:** [More Information Needed]
+- **Developed by:** [Nakato](https://huggingface.co/nakato-nk)
 - **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]
-
-### Model Sources [optional]
+- **Model type:** chat
+- **Language(s) (NLP):** Español, Ingles
+- **License:** [cc-by-nc-sa-4.0](https://creativecommons.org/licenses/by-nc-sa/4.0/deed.en)
+- **Finetuned from model:** [unsloth/Llama-3.2-1B-Instruct](https://huggingface.co/unsloth/Llama-3.2-1B-Instruct)
+- **Model Release Date**: Marzo 22, 2025
 
+### Model Sources
 <!-- Provide the basic links for the model. -->
 
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
-
-## Uses
-
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-
-### Direct Use
-
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-
-[More Information Needed]
-
-### Downstream Use [optional]
-
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-
-[More Information Needed]
-
-### Out-of-Scope Use
-
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-
-[More Information Needed]
+- **Repository:** [RaMem Github](https://github.com/Deep-Sphere-AI/RaMem)
 
 ## Bias, Risks, and Limitations
 
 <!-- This section is meant to convey both technical and sociotechnical limitations. -->
-
-[More Information Needed]
+Al ser un Modelo pequeño, tiene algunas limitaciones, ya sea en la coherencia de las respuestas, alucinaciones, ventana de contexto pequeña, etc. Las capacidades de este modelo vienen potenciadas por el RAG, sin ellas es solo un modelo de 1B de parametros finetuneado para español.
 
 ### Recommendations
 
 <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+Es un modelo pequeño, es óptimo para correr en hardware limitado, lo que hace que la mayoría de personas pueden usarlo. Se recomiendo usar la plantilla de chat predefinida y no pasar de la ventana de contexto de 2048 tokens. Tampoco usarlo de forma profesional como ChatGPT.
 
 ## How to Get Started with the Model
 
-Use the code below to get started with the model.
-
-[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]
+El dataset usado es una mezcla entre el ya conocido OASST2 y un datset sintétic elaborado con la ayuda de modelos como DeepSeek y ChatGPT, todos los ejemplos tienen un system prompt. Y se aplicó el chat template de LLama 3.1
 
 ### 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]
-
+#### Preprocessing  
+Se ha usado el chat template proporcionado por Meta, para LLama 3.1, el dataset ya se encuentra en un formato que permite aplicar el chat tempalte.
+Antes de tokenizar los datos se hagrega un sistem prompt en caso las conversaciones no lo tengan. Además se ha cambiado en todo el dataset las referencias sobre el nombre del modelo por RaMem.
 
 #### 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]
-
-## 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]
+- **max_seq_length**: 4096 (RoPE scaling)
+- **lora_alpha**: 16
+- **lora_dropout**: 0
+- **epochs**: 1
+- **learning_rate**: 2e-4
+- **per_device_train_batch_size**: 2
+- **gradient_accumulation_steps**: 4
+-
 
-[More Information Needed]
+## Model Card Authors
 
-## Model Card Contact
+- Nakato
 
-[More Information Needed]
 ### Framework versions
 
-- PEFT 0.14.0
+- PEFT 0.14.0
+- unsloth 2025.3.17