commit4

README.md

@@ -2,91 +2,62 @@
 language:
 - en
 license: mit
-tags:
-- maba
-- maba-v1.5
-- recurrent
-- dgda
-- linear-attention
-- sparse-attention
-- maba-sa
-- mla
-- nope
 pipeline_tag: text-generation
 ---
 
-<p align="center">
-  <img src="https://huggingface.co/AndrewThompson1233/maba-1.5-103m/resolve/main/assets/logo.svg" width="160" alt="Maba Logo" />
-</p>
+<p align="center"><img src="assets/logo.svg" width="160" alt="Logo" /></p>
 
 # Maba v1.5 (103.5M)
 
-Trained checkpoint built on the **[AndrewThompson1233/maba-v1.5-exp-architecture](https://huggingface.co/AndrewThompson1233/maba-v1.5-exp-architecture)** reference architecture.
+Trained checkpoint built on [AndrewThompson1233/maba-v1.5-exp-architecture](https://huggingface.co/AndrewThompson1233/maba-v1.5-exp-architecture).
 
 * Base Architecture: [AndrewThompson1233/maba-v1.5-exp-architecture](https://huggingface.co/AndrewThompson1233/maba-v1.5-exp-architecture)
-* Parameters: **103,520,911 (103.5M)**
-* Core Computation Ratio: **95.21%** (4.30% Vocab Tax)
-* Macro-Stack: **3:1** (15 DGDA Recurrence : 5 MABA-SA Dynamic Sparse Attention)
+* Parameters: **103.5M**
+* Core Ratio: **95.21%** (4.30% Vocab Tax)
+* Macro-Stack: **3:1** (15 DGDA Recurrence : 5 MABA-SA Attention)
 * Positional Encoding: **Strict NoPE** (0 parameters)
-* Training Corpus: 3,044 dialogue pairs on NVIDIA L4 (bfloat16)
-
----
+* Trained on: 3,044 dialogues on NVIDIA L4 (bfloat16)
 
 ## Empirical Benchmark vs Qwen3.8-Flash-Next (101.7M)
 
-Evaluated under identical training budgets (3,044 dialogues, 15 epochs, bfloat16, NVIDIA L4):
-
-| Metric | Maba v1.5-exp | Qwen3.8-Flash-Next | Delta / Advantage |
+| Metric | Maba v1.5-exp | Qwen3.8-Flash-Next | Delta |
 | :--- | :---: | :---: | :---: |
-| **Parameters** | **103,520,911 (103.5M)** | 101,701,120 (101.7M) | 0.2% parity |
-| **Architecture** | **75% DGDA + 25% MABA-SA** | 75% GDN + 25% QSA + MoE | Cyclic 3:1 |
-| **Positional Encoding** | **Strict NoPE (0 params)** | 25% Partial RoPE | Zero positional overhead |
-| **Contrastive Retrieval (MCQ)** | **87.5% (7/8)** | 75.0% (6/8) | **+12.5% accuracy** |
-| **Validation Loss** | **0.0697** | 0.0778 | **-10.4% entropy** |
-| **Validation Perplexity (PPL)** | **1.07** | 1.08 | **Maba wins** |
-| **Decode Throughput (L4)** | **7.0 tok/s** | 5.5 tok/s | **+27.3% faster generation** |
-
-<p align="center">
-  <img src="https://huggingface.co/AndrewThompson1233/maba-1.5-103m/resolve/main/assets/empirical_benchmark.svg" width="920" alt="Maba v1.5 Empirical Benchmark" />
-</p>
+| **Retrieval Accuracy (MCQ)** | **87.5% (7/8)** | 75.0% (6/8) | **+12.5%** |
+| **Validation Loss** | **0.0697** | 0.0778 | **-10.4%** |
+| **Perplexity (PPL)** | **1.07** | 1.08 | **Maba wins** |
+| **Decoding Speed (L4)** | **7.0 tok/s** | 5.5 tok/s | **+27.3% faster** |
 
----
-
-## Attention Ablation Proof
+<p align="center"><img src="assets/architecture_comparison.svg" width="920" alt="Benchmark" /></p>
 
-Empirical demonstration of the contribution of the 25% MABA-SA dynamic sparse attention layers against a pure linear recurrent baseline on the exact same checkpoint weights:
+## How to Use
 
-| Model Variant | Attention Mechanism | Validation Loss | Perplexity (PPL) | Error Reduction |
-| :--- | :---: | :---: | :---: | :---: |
-| **Pure DGDA (Ablation)** | None (100% Linear Recurrence) | 3.9360 | 51.21 | Baseline |
-| **Qwen3.8-Flash-Next** | QSA (GQA + Micro-block Indexer) | 3.8772 | 48.29 | -5.7% vs Recurrence |
-| **Maba v1.5 Full** | **MABA-SA (MLA + Top-32 + HCA)** | **3.5903** | **36.24** | **-29.2% error drop** |
+```python
+import sys, subprocess, torch
+from huggingface_hub import hf_hub_download
+from safetensors.torch import load_file
+from transformers import AutoTokenizer
 
----
+tokenizer = AutoTokenizer.from_pretrained('gpt2')
+if 'maba-v1.5-exp-architecture' not in sys.path:
+    subprocess.run(['git', 'clone', 'https://huggingface.co/AndrewThompson1233/maba-v1.5-exp-architecture'], check=False)
+    sys.path.insert(0, 'maba-v1.5-exp-architecture')
 
-## Needle-In-A-Haystack & Centroid Retrieval (512 to 4096 Tokens)
+from maba_sparse.config import MabaSparseConfig
+from maba_sparse.model import MabaSparseForCausalLM
 
-| Context Length | Needle Position | Needle Block | DG-Indexer (Hybrid Mean+Max) | Standard Pure Mean Pooling |
-| :---: | :---: | :---: | :---: | :---: |
-| **512 tokens** | 51 (10%) | Block #0 | **Retrieved (Top-32)** | Retrieved |
-| **512 tokens** | 256 (50%) | Block #4 | **Retrieved (Top-32)** | Retrieved |
-| **512 tokens** | 460 (90%) | Block #7 | **Retrieved (Top-32)** | Retrieved |
-| **1024 tokens** | 102 (10%) | Block #1 | **Retrieved (Top-32)** | Retrieved |
-| **1024 tokens** | 512 (50%) | Block #8 | **Retrieved (Top-32)** | Retrieved |
-| **1024 tokens** | 921 (90%) | Block #14 | **Retrieved (Top-32)** | Retrieved |
-| **2048 tokens** | 204 (10%) | Block #3 | **Retrieved (Top-32)** | Retrieved |
-| **2048 tokens** | 1024 (50%) | Block #16 | **Retrieved (Top-32)** | Retrieved |
-| **2048 tokens** | 1843 (90%) | Block #28 | **Retrieved (Top-32)** | Retrieved |
-| **4096 tokens** | 2048 (50% Lost-in-Middle) | Block #32 | **Retrieved (Top-32)** | **Diluted to 0.0 (Failed)** |
-| **4096 tokens** | 3686 (90%) | Block #57 | **Retrieved (Top-32)** | Retrieved |
+device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+cfg = MabaSparseConfig(vocab_size=len(tokenizer), dim=640, d_emb=128, intermediate_size=1248, n_layers=20)
+model = MabaSparseForCausalLM(cfg).to(device)
 
----
+weights_path = hf_hub_download(repo_id='AndrewThompson1233/maba-1.5-103m', filename='model.safetensors')
+model.load_state_dict(load_file(weights_path))
+model.eval()
 
-## KV-Cache Footprint at 4k Context
+prompt = 'User: tell me a joke\\nAssistant: '
+ids = tokenizer.encode(prompt, return_tensors='pt').to(device)
+with torch.no_grad():
+    with torch.amp.autocast('cuda', dtype=torch.bfloat16):
+        out = model.generate(ids, max_new_tokens=35, temperature=0.0)
 
-| Context Length | Dense Attention (Baseline) | Qwen3.8-Flash-Next | Maba v1.5 (MLA + Top-32) | Memory Reduction vs Dense |
-| :---: | :---: | :---: | :---: | :---: |
-| **512 tokens** | 25.00 MB | 1.00 MB | **0.62 MB** | **-97.5%** |
-| **1,024 tokens** | 50.00 MB | 2.00 MB | **1.25 MB** | **-97.5%** |
-| **2,048 tokens** | 100.00 MB | 4.00 MB | **2.50 MB** | **-97.5%** |
-| **4,096 tokens** | 200.00 MB | 8.00 MB | **2.50 MB** | **-98.8%** |
+print(tokenizer.decode(out[0][ids.shape[1]:], skip_special_tokens=True).strip())
+```