Upload MODEL_METHODOLOGY.md with huggingface_hub

MODEL_METHODOLOGY.md

@@ -0,0 +1,184 @@
+---
+tags:
+- 7b
+- Chinese
+- English
+- android
+- apple-silicon
+- code
+- compensation-lora
+- continuum
+- distillation
+- edge-inference
+- efficient
+- embedded
+- experiential-plasticity
+- forge-alloy
+- forged
+- general
+- general-purpose
+- head-pruning
+- iphone
+- llama-cpp
+- lm-studio
+- local-inference
+- lora
+- macbook
+- mobile
+- neural-plasticity
+- ollama
+- on-device
+- optimized
+- pruned
+- qwen
+- qwen2.5
+- raspberry-pi
+- sentinel-ai
+- text-generation
+- validation-artifact
+- versatile
+base_model: Qwen/Qwen2.5-Coder-7B
+pipeline_tag: text-generation
+license: apache-2.0
+---
+
+# 12% Pruned, 61.0 HUMANEVAL (base 62.2)
+
+**Qwen2.5-Coder-7B** forged through Experiential Plasticity and recovered to within calibration tolerance of the unmodified base via KL-distillation compensation LoRA.
+
+- **HUMANEVAL**: 61.0 (base 62.2, Δ -1.2)
+- **HUMANEVAL+PLUS**: 53.0 (base 53.7, Δ -0.7)
+
+
+<p align="center">
+<a href="https://cambriantech.github.io/forge-alloy/verify/#c7be31309161f9ca">
+<img src="alloy-qr.png" alt="Verify Chain of Custody" width="160"/>
+</a>
+</p>
+
+<p align="center">
+<a href="https://cambriantech.github.io/forge-alloy/verify/#c7be31309161f9ca"><b>Every claim on this card is verified</b></a><br>
+<b>Trust: self-attested</b> · 2 benchmarks · 1 device tested<br>
+<a href="https://github.com/CambrianTech/forge-alloy">ForgeAlloy</a> chain of custody · <a href="v2-7b-coder-compensated.alloy.json">Download alloy</a> · Merkle-chained
+</p>
+
+---
+
+## About this model
+
+Methodology validation artifact for the v2 forge pipeline + KL-distillation compensation LoRA. Demonstrates that aggressive head pruning + activation-metric importance + pad-mode defrag, when paired with output-distribution distillation against the unmodified teacher, recovers near-base HumanEval capability (61.0 vs 62.2 base, within calibration tolerance). This is the empirical anchor for PLASTICITY-COMPACTION §4.1.3.3 and the loss-function ablation that closes the §4.1.3.2 PPL/HumanEval disconnect. NOT a Pareto improvement over the unmodified base 7B at any single VRAM tier — published as proof that the methodology stack works end-to-end, in preparation for the Qwen3.5-35B-A3B and 397B-A17B forges where the pruning dimension actually wins.
+
+## The Journey
+
+This artifact is the punchline of a four-run experimental sequence on the same base model. The first run scored **50.0**; the final run scored **61.0**. Each run between them isolated a single variable, and each result narrowed the design space to the structural fix that recovered near-base capability.
+
+| Run | Configuration | HumanEval pass@1 |
+|---|---|---|
+| 1 | broken global-flat L2-weight | **50.0** |
+| 2 | layer-normalized activation, 1-cycle 500-step | **54.9** |
+| 3 | layer-normalized activation, 3-cycle (ablation) | **46.3** |
+| 4 | 1-cycle + KL compensation LoRA | **61.0** |
+
+## Loss Function Ablation
+
+The compensation LoRA was run twice with identical configuration, varying only the distillation loss. The result is a substantive methodology finding in its own right:
+
+| Distillation loss | HumanEval | HumanEval+ | Outcome |
+|---|---|---|---|
+| `mse_hidden` | **0.0** | **0.0** | degenerate fixed point — model collapsed to outputting '0' |
+| `kl_logits` | **61.0** | **53.0** | near-base recovery within calibration tolerance |
+
+MSE-on-hidden-states has a degenerate fixed point: the student can satisfy the loss by collapsing some downstream computation, regardless of whether the hidden states encode useful information. KL-on-output-logits has none, because matching the teacher's output distribution directly constrains task-level behavior. **For autoregressive language models, distillation must operate at the output layer, not at intermediate residual streams.**
+
+
+## Benchmarks
+
+| Benchmark | Score | Base | Δ | Verified |
+|---|---|---|---|---|
+| **humaneval** | **61.0** | 62.2 | -1.2 | ✅ Result hash |
+| **humaneval_plus** | **53.0** | 53.7 | -0.7 | ✅ Result hash |
+
+
+## What Changed (Base → Forged)
+
+| | Base | Forged | Delta |
+|---|---|---|---|
+| **Pruning** | None | 12% heads (activation-magnitude) | **-12%** params ✅ |
+| **LoRA** | None | rank=? |  |
+| **Pipeline** | | prune → lora → lora → eval | 1 cycles |
+
+## Runs On
+
+| Device | Format | Size | Speed |
+|--------|--------|------|-------|
+| **NVIDIA GeForce RTX 5090** | fp16 | — | Verified |
+| MacBook Pro 32GB | fp16 | 8.0GB | Expected |
+| MacBook Air 16GB | Q8_0 | ~4.0GB | Expected |
+| MacBook Air 8GB | Q4_K_M | ~2.5GB | Expected |
+| iPhone / Android | Q4_K_M | ~2.5GB | Expected |
+
+## Quick Start
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = AutoModelForCausalLM.from_pretrained("continuum-ai/v2-7b-coder-compensated",
+    torch_dtype="auto", device_map="auto")
+tokenizer = AutoTokenizer.from_pretrained("continuum-ai/v2-7b-coder-compensated")
+
+inputs = tokenizer("def merge_sort(arr):", return_tensors="pt").to(model.device)
+output = model.generate(**inputs, max_new_tokens=200)
+print(tokenizer.decode(output[0], skip_special_tokens=True))
+```
+
+
+## How It Was Made
+
+```
+prune → lora → lora → eval (1 cycles)
+```
+
+- **Pruning**: 12% heads via `activation-magnitude`, layer-normalized, pad-mode defrag
+  > Layer-normalized activation-magnitude head importance (PLASTICITY-COMPACTION §4.1.3.1 fix). Pad-mode defrag preserves the q_proj invariant num_q_heads*head_dim==hidden_size so the artifact loads in llama.cpp (Finding 6 fix from VALIDATED-TENSOR-SURGERY).
+- **lora**: rank ?, 500 steps
+  > Single-cycle code-domain LoRA fine-tuning on the pruned student. 1-cycle ablation chosen because the 3-cycle multi-cycle test surfaced the §4.1.3.2 PPL/HumanEval disconnect (54.9 → 46.3 across cycles).
+- **compensation-lora**: rank 16, 500 steps, `kl_logits` distillation against `Qwen/Qwen2.5-Coder-7B`
+  > PLASTICITY-COMPACTION §4.1.3.3. KL divergence on output logits is the structural fix for the §4.1.3.2 disconnect. Loss-function ablation: MSE-on-hidden-states collapsed the model to 0.0 (degenerate fixed point); KL-on-logits recovered to 61.0. LoRA adapter merged into student weights at save time so inference-time VRAM and tokens/sec are unchanged from the un-compensated student.
+- **Calibrated evaluation**: anchored against `Qwen2.5-Coder-7B` (published 61.6, measured 62.2, ±3.0pt tolerance)
+  > All HumanEval numbers are anchor-calibrated against the unmodified Qwen2.5-Coder-7B base measured on the same hardware/pipeline in the same run. Hard-fail tolerance: ±3.0 points. Anchor delta: +0.6/+0.7 vs Qwen-published 61.6/53.0, deterministic across 6+ independent runs.
+- **Hardware**: NVIDIA GeForce RTX 5090
+- **Forge tool**: [Continuum](https://github.com/CambrianTech/continuum) Factory + [sentinel-ai](https://github.com/CambrianTech/sentinel-ai)
+## Limitations
+
+- This model is currently a methodology demonstration rather than a Pareto-optimal artifact at any specific hardware tier. For production code workloads on smaller hardware, the unmodified Qwen2.5-Coder-7B at standard quantization (Q4_K_M / Q5_K_M / Q8_0) may be a better fit pending the larger Qwen3.5+ forges that exercise the pruning dimension where this methodology actually wins.
+- Validated on HumanEval / HumanEval+ for English-language Python code completion. Performance on other programming languages, code paradigms (functional, embedded, kernel), or code-adjacent domains (SQL, regex, shell) has not been measured.
+- Ships as fp16 only. GGUF quantization tiers (Q5_K_S / Q3_K_M / Q2_K) are not yet published for this artifact; the per-tier comparison from the development log showed base+quant dominates v2+quant at every VRAM tier on the same 7B base, which is why the methodology validation here uses fp16 and the production GGUF publishes are reserved for the Qwen3.5+ forges where the dimension flips.
+- Vision modality not yet wired in. The Continuum sensory architecture treats vision as first-class for personas, but this 7B coder artifact is text-only.
+
+
+## Chain of Custody
+
+Scan the QR or [verify online](https://cambriantech.github.io/forge-alloy/verify/#c7be31309161f9ca). Download the [alloy file](v2-7b-coder-compensated.alloy.json) to verify independently.
+
+| What | Proof |
+|------|-------|
+| Forged on | NVIDIA GeForce RTX 5090, ? |
+| Published | [huggingface](https://huggingface.co/continuum-ai/v2-7b-coder-compensated) — 2026-04-08T05:01:40.446154+00:00 |
+| Trust level | [`self-attested`](https://github.com/CambrianTech/forge-alloy/blob/main/docs/ATTESTATION.md) |
+| Spec | [ForgeAlloy](https://github.com/CambrianTech/forge-alloy) — Rust/Python/TypeScript |
+
+## Make Your Own
+
+Forged with [Continuum](https://github.com/CambrianTech/continuum) — a distributed AI world that runs on your hardware.
+
+<p align="center">
+<a href="https://github.com/CambrianTech/continuum"><img src="https://raw.githubusercontent.com/CambrianTech/continuum/main/docs/images/factory.png" alt="Continuum Model Factory" width="400"/></a>
+</p>
+
+The Factory configurator lets you design and forge custom models visually — context extension, pruning, LoRA, quantization, vision/audio modalities. Pick your target devices, the system figures out what fits.
+
+[GitHub](https://github.com/CambrianTech/continuum) · [All Models](https://huggingface.co/continuum-ai) · [Forge-Alloy](https://github.com/CambrianTech/forge-alloy)
+
+## License
+
+apache-2.0