hm-016: framework-only public 7B (clean distill, eval 0.975/OOD 0.000)

.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 +1,114 @@
+---
+base_model: unsloth/Qwen2.5-7B-Instruct-bnb-4bit
+library_name: peft
+pipeline_tag: text-generation
+license: apache-2.0
+tags:
+- lora
+- qlora
+- sft
+- unsloth
+- trl
+- hammerstein
+- strategic-reasoning
+---
+
+# Hammerstein-7B Framework — a small, opinionated strategic-reasoning model
+
+A QLoRA adapter on `Qwen2.5-7B-Instruct` that bakes the
+[Hammerstein framework](https://github.com/lerugray/hammerstein) into the
+weights via behavior cloning. Load base + adapter, run inference **with no
+system prompt at all**, and you get framework-correct strategic reasoning:
+it names which failure quadrant a plan sits in (clever-lazy /
+clever-industrious / stupid-industrious / stupid-lazy), pairs claims with
+counter-observations, and proposes structural fixes over discipline fixes.
+
+This is the **framework-only public artifact** (refreshed 2026-06-05). It is
+trained on a framework-corpus distillation **with zero personal data** — the
+training set was deterministically scrubbed and passed an adversarial
+multi-agent privacy sweep before release. (Ongoing personal-corpus
+fine-tuning of the author's daily-driver continues privately and is not
+published here.)
+
+## What it does that frontier assistants are tuned *not* to do
+
+The framework deliberately reinforces three behaviors the big labs train
+toward agreeableness and away from:
+
+- **Refusal-with-pathway** — when the right answer is "don't do this," it
+  says so, and surfaces what *would* unblock a yes, instead of a flat no or
+  a reluctant yes.
+- **Hold-your-ground** — it does not sycophantically fold when you push back
+  with confidence but no new evidence. It restates the structural reason and
+  tells you exactly what evidence would change its call.
+- **Refuse stupid-industrious** — it declines to validate a confidently-stated
+  plan that works hard in the wrong direction; it names the quadrant and
+  offers a verification gate + structural alternative.
+
+## Training data (framework-only, 1,994 pairs)
+
+| Source | Pairs | What |
+|---|---|---|
+| Strategic (scrubbed v3a corpus) | 1,708 | audit-this-plan / scope-this-idea / is-this-worth-doing / what-should-we-do-next / review-from-different-angle, across 12 generic domains |
+| Unique-behavior reinforcement | 72 | the three doctrine behaviors above (24 each) |
+| Off-domain instruction-following | 214 | suppresses catastrophic forgetting (keeps general competence) |
+
+Teacher: Qwen3.6-plus running the Hammerstein framework prompt (no corpus
+retrieval, neutralized persona — clean by construction). Behavior-cloning
+frame: **no system prompt in the training targets** — the framework is what
+the student learns to bake in.
+
+## Eval (framework-discipline benchmark, 2026-06-05)
+
+Structural framework-correctness on 40 held-out strategic prompts
+(higher = more framework-correct), and an out-of-domain forgetting check
+on 30 prompts (framework-vocab leakage into off-domain answers; lower =
+healthier):
+
+| Condition | Strategic (n=40) | OOD leakage (n=30) |
+|---|---|---|
+| **student** (this adapter, **no system prompt**) | **0.975** | **0.000** |
+| ablation (base + framework system prompt) | 0.675 | 0.783 |
+| vanilla (base Qwen2.5-7B alone) | 0.081 | 0.000 |
+
+**Adapter wins (Δ=+0.300 vs the prompt-only ablation) — the framework
+lives in the weights, not just a runtime prompt.** OOD leakage is 0.000:
+the distillation adds framework discipline with no measurable
+catastrophic forgetting. Note the prompt-only ablation actually *leaks*
+framework vocabulary into off-domain answers (0.783) where the distilled
+student does not — the student fires the framework when the task calls
+for it and stays quiet when it doesn't.
+
+The framework-fidelity axis is partly tautological (the rubric rewards
+framework vocabulary by design); the load-bearing signal is that the
+distillation carries the *discipline* into 7B weights with no runtime
+scaffolding, and does not wreck general competence (forgetting ≈ 0).
+
+## Usage
+
+```python
+from peft import PeftModel
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+base = "Qwen/Qwen2.5-7B-Instruct"
+tok = AutoTokenizer.from_pretrained(base)
+model = AutoModelForCausalLM.from_pretrained(base, device_map="auto")
+model = PeftModel.from_pretrained(model, "lerugray/hammerstein-7b-framework")
+
+msgs = [{"role": "user", "content": "Audit this plan: rewrite our API gateway from scratch in Rust to fix latency."}]
+ids = tok.apply_chat_template(msgs, add_generation_prompt=True, return_tensors="pt").to(model.device)
+print(tok.decode(model.generate(ids, max_new_tokens=600)[0][ids.shape[1]:], skip_special_tokens=True))
+```
+
+No system prompt needed. Runs locally on an 8 GB GPU at zero per-call cost.
+
+## What this is not
+
+Not a general-purpose frontier replacement. It is tuned for framework-shaped
+strategic-reasoning tasks; generalization to neutral benchmarks (math, code,
+long-context) is untested. The **framework is the IP**; this adapter is the
+portability proof — a small owned model that holds an opinionated reasoning
+doctrine you can run yourself.
+
+Built alongside [hammerstein.ai](https://hammerstein.ai). Framework + corpus:
+[github.com/lerugray/hammerstein](https://github.com/lerugray/hammerstein).

adapter_config.json

@@ -0,0 +1,52 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen2ForCausalLM",
+    "parent_library": "transformers.models.qwen2.modeling_qwen2",
+    "unsloth_fixed": true
+  },
+  "base_model_name_or_path": "unsloth/Qwen2.5-7B-Instruct-bnb-4bit",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "gate_proj",
+    "o_proj",
+    "down_proj",
+    "up_proj",
+    "k_proj",
+    "q_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5267ff2a51a736111bf68a6aa4e5005903df9797e54e2dd3415f3918d96a0033
+size 323014168

chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd5948af71b4f56cf697f7580814c7ce8b80595ef985544efcacf716126a2e31
+size 11422356

tokenizer_config.json

@@ -0,0 +1,201 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "is_local": false,
+  "model_max_length": 32768,
+  "pad_token": "<|PAD_TOKEN|>",
+  "padding_side": "left",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<|PAD_TOKEN|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    }
+  }
+}