Dystrio Sculpt default tier of gemma 2 2b it

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,139 @@
+---
+license: apache-2.0
+library_name: transformers
+pipeline_tag: text-generation
+language:
+  - en
+base_model: google/gemma-2-2b-it
+tags:
+  - dystrio
+  - sculpt
+  - pruned
+  - compressed
+  - efficient
+  - dense
+  - runtime-agnostic
+  - no-custom-kernels
+  - hf-drop-in
+  - drop-in-replacement
+  - smaller
+  - faster
+  - gemma
+datasets:
+  - wikitext
+model-index:
+  - name: Dystrio Sculpt (gemma-2-2b-it Default)
+    results:
+      - task:
+          type: text-generation
+        dataset:
+          name: WikiText-103 (validation)
+          type: wikitext
+        metrics:
+          - name: perplexity
+            type: perplexity
+            value: 20.5854
+          - name: ppl_ratio
+            type: ppl_ratio
+            value: 0.7985
+---
+
+# dystrio/gemma-2-2b-it-sculpt-default
+
+> **9% smaller, quality improved (0.7985x PPL), drop-in replacement. No custom kernels. No runtime changes.**
+
+Dystrio Sculpt structurally compresses transformer models, producing dense models that load with standard `transformers` — no custom code, no new ops, no deployment friction.
+
+This is the **Default** tier of [gemma 2 2b it](https://huggingface.co/google/gemma-2-2b-it).
+
+## Quick Start
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = AutoModelForCausalLM.from_pretrained("dystrio/gemma-2-2b-it-sculpt-default", torch_dtype="bfloat16", device_map="auto")
+tokenizer = AutoTokenizer.from_pretrained("dystrio/gemma-2-2b-it-sculpt-default")
+
+inputs = tokenizer("The future of AI inference is", return_tensors="pt").to(model.device)
+outputs = model.generate(**inputs, max_new_tokens=100)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+```
+
+## Benchmark Results
+
+All tiers compiled from [gemma 2 2b it](https://huggingface.co/google/gemma-2-2b-it) on A100 80GB, bf16:
+
+| Model | PPL | PPL Ratio | Weights (GB) | Chat Prefill TPS | RAG TTFT p95 (ms) | Decode TPS |
+|-------|-----|-----------|-------------|------------------|-------------------|------------|
+| **Baseline** | 25.7807 | 1.0 | 4.869591 | 21611.9 | 70.251 | 59.6 |
+| **sculpt-default** | 20.5854 | 0.7985 | 4.441124 | 23065.3 | 69.007 | 60.0 |
+| **sculpt-production** | 22.4118 | 0.8693 | 4.226891 | 23404.0 | 66.554 | 60.7 |
+| **sculpt-throughput** | 29.8372 | 1.1573 | 3.969811 | 24330.1 | 64.529 | 59.3 |
+| **sculpt-experimental** | 48.9699 | 1.8995 | 3.412804 | 26496.2 | 60.97 | 59.5 |
+
+### Key Metrics (this model)
+
+| Metric | Value |
+|--------|-------|
+| **Weights memory** | 4.441124 GB (9% smaller) |
+| **PPL ratio** | 0.7985 |
+| **Chat prefill TPS** | 23065.3 (+7%) |
+| **RAG TTFT p95** | 69.007 ms (-2%) |
+| **Decode TPS** | 60.0 (flat) |
+| **Parameters** | 2.38B |
+
+## All Sculpt Tiers
+
+| Tier | HuggingFace | Size | PPL Ratio | Use Case |
+|------|-------------|------|-----------|----------|
+| default | [dystrio/gemma-2-2b-it-sculpt-default](https://huggingface.co/dystrio/gemma-2-2b-it-sculpt-default) 👈 **this model** | 4.441124 GB | 0.7985 | Zero-regret: quality preserved, smaller footprint |
+| production | [dystrio/gemma-2-2b-it-sculpt-production](https://huggingface.co/dystrio/gemma-2-2b-it-sculpt-production) | 4.226891 GB | 0.8693 | Practical savings with modest quality tradeoff |
+| throughput | [dystrio/gemma-2-2b-it-sculpt-throughput](https://huggingface.co/dystrio/gemma-2-2b-it-sculpt-throughput) | 3.969811 GB | 1.1573 | Maximum usable compression for speed/edge |
+| experimental | [dystrio/gemma-2-2b-it-sculpt-experimental](https://huggingface.co/dystrio/gemma-2-2b-it-sculpt-experimental) | 3.412804 GB | 1.8995 | Boundary exploration, maximum structural compression |
+
+## What is Dystrio Sculpt?
+
+Dystrio Sculpt compiles transformer models into smaller, faster variants. Output models:
+
+- Are **dense** (not sparse) — standard architecture, fewer parameters
+- Load with **standard HuggingFace Transformers** — no custom code needed
+- Require **no custom kernels** and **no runtime changes**
+- Work as a one-step compile before deployment
+- Stack with quantization (AWQ, GPTQ, GGUF) for compound savings
+
+## Compatibility
+
+- ✅ HuggingFace Transformers
+- ✅ vLLM
+- ✅ TGI (Text Generation Inference)
+- ✅ llama.cpp / GGUF conversion
+- ✅ AWQ / GPTQ quantization
+- ✅ Any framework that loads standard safetensors
+
+## Benchmark Environment
+
+- **GPU**: NVIDIA A100-SXM4-80GB
+- **dtype**: bf16
+- **Torch**: 2.10.0+cu128
+- **Transformers**: 5.3.0
+- **Deterministic**: True
+- Single-GPU, standard HuggingFace Transformers, no custom kernels.
+
+## Metric Definitions
+
+- **PPL ratio**: WikiText-103 perplexity relative to baseline. <1.0 = quality improved.
+- **Prefill TPS**: Tokens per second during prompt encoding (higher = faster).
+- **TTFT p95**: Time to first token at 95th percentile (lower = faster).
+- **Decode TPS**: Tokens per second during generation (higher = faster).
+- **Weights (GB)**: Model parameter memory (deterministic, runtime-independent).
+
+## Citation
+
+```bibtex
+@misc{dystrio_sculpt_2026,
+  title={Dystrio Sculpt: Structural Compilation for Transformer LLMs},
+  author={Dystrio},
+  year={2026},
+  url={https://huggingface.co/dystrio}
+}
+```

chat_template.jinja

@@ -0,0 +1,4 @@
+{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '
+' + message['content'] | trim + '<end_of_turn>
+' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model
+'}}{% endif %}

config.json

@@ -0,0 +1,68 @@
+{
+  "architectures": [
+    "Gemma2ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "attn_logit_softcapping": 50.0,
+  "bos_token_id": 2,
+  "cache_implementation": "hybrid",
+  "dtype": "bfloat16",
+  "eos_token_id": [
+    1,
+    107
+  ],
+  "final_logit_softcapping": 30.0,
+  "head_dim": 256,
+  "hidden_act": "gelu_pytorch_tanh",
+  "hidden_activation": "gelu_pytorch_tanh",
+  "hidden_size": 2304,
+  "initializer_range": 0.02,
+  "intermediate_size": 7936,
+  "layer_types": [
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 8192,
+  "model_type": "gemma2",
+  "num_attention_heads": 8,
+  "num_hidden_layers": 26,
+  "num_key_value_heads": 4,
+  "pad_token_id": 0,
+  "query_pre_attn_scalar": 256,
+  "rms_norm_eps": 1e-06,
+  "rope_parameters": {
+    "rope_theta": 10000.0,
+    "rope_type": "default"
+  },
+  "sliding_window": 4096,
+  "tie_word_embeddings": true,
+  "transformers_version": "5.3.0",
+  "use_bidirectional_attention": null,
+  "use_cache": true,
+  "vocab_size": 256000
+}

generation_config.json

@@ -0,0 +1,11 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "cache_implementation": "hybrid",
+  "eos_token_id": [
+    1,
+    107
+  ],
+  "pad_token_id": 0,
+  "transformers_version": "5.3.0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76baecdc7f408517381ba65508f9b9577bf2f23d3d41a7301d2ad641105b4242
+size 4768654792

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6994501c3d2c96eda15f5cb4d75ffd2cb41fa8e53e9cafc73db6adae158bf147
+size 34362970

tokenizer_config.json

@@ -0,0 +1,19 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<bos>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "extra_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "is_local": false,
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}