seedboxai
/

KafkaLM-15B

@@ -1,190 +1,80 @@
 ---
 library_name: transformers
-tags: []
 ---
-# Model Card for Model ID
-<!-- Provide a quick summary of what the model is/does. -->
-## Model Details
-### Model Description
-<!-- Provide a longer summary of what this model is. -->
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
-- **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]
-### Model Sources [optional]
-<!-- 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]
-## Bias, Risks, and Limitations
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-[More Information Needed]
-### 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.
-## 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]
-### 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]
-## 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]

 ---
 library_name: transformers
+tags:
+- pruning
+- distillation
+- sparsity‑2:4
+license: apache-2.0
+language:
+- en
+- de
+- fr
+- es
+- it
+- pt
+base_model:
+- doubledsbv/KafkaLM-15B-Base
+pipeline_tag: text-generation
 ---
+# Model Description
+**KafkaLM‑15B‑Base** is a 15‑billion‑parameter, sparsity‑aware language model distilled from *Mistral‑Small‑24B‑Base‑2501*.
+This experimental model was created in three stages:
+| Stage | What we did | Why it matters |
+|-------|-------------|----------------|
+| **1. SimplePrune** | Applied a hierarchical, hardware‑aware pruning pipeline that combines block‑, channel‑ and 2:4 structured sparsity (≈ 37.5 % parameter reduction) | Slashes memory footprint while minimizing perplexity degradation |
+| **2. Teacher calibration** | Briefly fine‑tuned the unpruned 24 B teacher on a 10 B‑token multilingual European corpus on a AMD M300A cluster | Produces stable logits and hidden states for distillation  |
+| **3. Knowledge distillation** | Distilled the calibrated teacher into the pruned 15 B student using a **fused loss**:<br/>`L Pooled SquareHead + LKL + 0.25 * LCE` | Transfers teacher capabiities effectively with <15B tokens **(< 2 epochs)** on 64 MI300A nodes |
+**Key capabilities**
+* Balanced for both **multitask** and multilingual conversation and long context handling
+* Structured **2:4 sparsity** → runs up to **40 % faster** on sparsity‑aware kernels
+* Distilled on a combination of multilingual pretraining and synthetic data
+* Training pipeline optimized for unified‑memory GPUs (AMD MI300A) but runs on any CUDA / ROCm device
+---
+## Pruning Process
+**Pruning & Distillation Strategy — SimplePrune**
+Hardware‑aware, hierarchical pipeline. SimplePrune starts with coarse block‑level pruning and drills down to channel‑ and neuron‑level removals, finishing with 2 : 4 structured sparsity. This staged approach converts compression ratios into real memory‑bandwidth and latency gains.
+**Sensitivity‑guided selection**
+Each stage is driven by activation‑magnitude profiles and Hessian‑based importance scores captured asynchronously during training, allowing the framework to run inside the MI300A’s 512 GB unified memory without OOM interruptions.
+**Two‑phase optimisation**
+A fast greedy pass prunes low‑impact blocks in MLP expansion layers, after which a **Tabu‑Search** meta‑heuristic explores cross‑layer combinations for a better global trade‑off between sparsity and perplexity/KL divergence.
+**Post‑pruning knowledge distillation**
+The pruned 15 B student is distilled from a calibrated 24 B teacher using a fused LSquareHead + KL + 0.25 · CE loss across 20 B multilingual tokens, restoring > 96 % of the original quality in ≤ 2 epochs on up to 64 MI300A nodes.
 ### Results
+Up to 40 % parameter reduction (24 B → 15 B) delivers 2× lower TTFT and ≈ 40 % higher tokens/s versus the uncompressed teacher while matching perplexity and divergence metrics—validating SimplePrune as an effective route to deploy KafkaLM in memory‑constrained, sparsity‑accelerated environments.
+| Metric | Mistral‑24B | **KafkaLM‑15B** | Δ |
+|--------|-------------|-----------------|---|
+| Time‑to‑First‑Token | 4.91 s | **2.46 s** | −50 % |
+| Prompts / s | 4.70 | **6.55** | +38 % |
+| Tokens / s | 579 | **812** | +40 % |
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/645ded34a45b4182d7f5c385/4rDhaeC-1GMj6KWbB27f9.png)
+### Training scalability (distillation run, MI300A cluster)
+| Nodes | Tokens / s | Speed‑up |
+|-------|------------|----------|
+| 4 | 1 461 | – |
+| 8 | 3 327 | 2.3 × |
+| 16 | 7 423 | 5.1 × |
+| 32 | 15 286 | 10.5 × |
+| 64 | 25 455 | 17.4 × |
+Near‑linear scaling thanks to sharded ZeRO‑3 + RCCL optimisations.
 ## More Information [optional]