Upload folder using huggingface_hub

README.md

@@ -1,102 +1,101 @@
----
-language: en
-license: mit
-tags:
-  - propagation-logic
-  - mechanism-first
-  - abstract-reasoning
-  - derivation-traces
-  - boundary-conditions
-datasets:
-  - ApplePiesFromScratch/dta-benchmark
-metrics:
-  - dta
----
-
-# MechanismBase — P / G → Q
-
-A 10M parameter transformer trained on derivation traces, not natural language.
-
-## What this is
-
-Standard language models learn statistical patterns over text.
-This model was trained on the **procedure** P / G → Q — explicit derivation
-traces showing closure analysis, fixed point detection, cycle structure
-identification, and forced boundary condition derivation.
-
-**The claim:** given any carrier V and gradient family Γ, the model can derive
-forced boundary conditions — what logic system the carrier implies, what
-fixed points exist, what cycle structure is forced.
-
-## Theory
-
-Propagation Logic v13 — SSRN Abstract ID: 6439258 (James Pugmire)
-
-The single primitive operator: `P / G → Q`
-
-A loaded pattern P propagates through gradient field G in context C to
-produce updated pattern Q. All of classical logic, fuzzy logic, arithmetic,
-calculus, and grammar fall out of different (V, Γ) choices.
-
-## Model
-
-- Architecture: Transformer decoder (custom, mechanism-aligned)
-- Parameters: 10.5M
-- Training tokens: ~1M (derivation traces)
-- Training epochs: 5
-
-## Benchmark: DTA (Derivation Trace Accuracy)
-
-The correct benchmark for this model is not BLiMP or MMLU.
-It is DTA — how accurately does the model predict forced boundary conditions
-on novel carriers?
-
-See: `ApplePiesFromScratch/dta-benchmark`
-
-| Model | DTA-Overall | DTA-Closure | DTA-FixedPts | DTA-Involution | DTA-Cycle |
-|-------|-------------|-------------|--------------|----------------|-----------|
-| MechanismBase (10M) | 62.5% | 80.0% | 70.0% | 70.0% | 60.0% |
-| Random baseline | 25% | 50% | 25% | 50% | 25% |
-| Engine (oracle) | 100% | 100% | 100% | 100% | 100% |
-
-
-## Usage
-
-```python
-# The model requires the pl/ library and engine.py from the repo
-# Clone: github.com/ApplePiesFromScratch/propagation-logic
-
-from model import MechanismBase, SmallConfig
-from tokenizers import Tokenizer
-import torch
-
-config = SmallConfig()
-model = MechanismBase(config)
-# Load weights from Hub (see full usage in repo)
-
-tokenizer = Tokenizer.from_file("mechanism_tokenizer/tokenizer.json")
-
-# Give the model a partial derivation trace
-partial = """DOMAIN: color_domain
-CARRIER: ['red', 'green', 'blue']
-GRADIENTS: ['complement', 'id']
-THETA: 1.0
----
-"""
-
-ids = torch.tensor(tokenizer.encode(partial).ids).unsqueeze(0)
-output = model.generate(ids, max_new_tokens=200, temperature=0.3)
-print(tokenizer.decode(output[0].tolist()))
-```
-
-## Training
-
-```
-python generate_data.py    # generates derivation trace corpus
-python tokenizer_train.py  # BPE tokenizer on corpus
-python train.py            # SmallConfig, ~30 min on RTX 4060 Ti
-```
-
-## Repository
-
-GitHub: [ApplePiesFromScratch/propagation-logic](https://github.com/ApplePiesFromScratch/propagation-logic)
+---
+language: en
+license: mit
+tags:
+  - propagation-logic
+  - mechanism-first
+  - abstract-reasoning
+  - derivation-traces
+  - boundary-conditions
+datasets:
+  - ApplePiesFromScratch/dta-benchmark
+metrics:
+  - dta
+---
+
+# MechanismBase — P / G → Q
+
+A 10M parameter transformer trained on derivation traces, not natural language.
+
+## What this is
+
+Standard language models learn statistical patterns over text.
+This model was trained on the **procedure** P / G → Q — explicit derivation
+traces showing closure analysis, fixed point detection, cycle structure
+identification, and forced boundary condition derivation.
+
+**The claim:** given any carrier V and gradient family Γ, the model can derive
+forced boundary conditions — what logic system the carrier implies, what
+fixed points exist, what cycle structure is forced.
+
+## Theory
+
+Propagation Logic v13 — SSRN Abstract ID: 6439258 (James Pugmire)
+
+The single primitive operator: `P / G → Q`
+
+A loaded pattern P propagates through gradient field G in context C to
+produce updated pattern Q. All of classical logic, fuzzy logic, arithmetic,
+calculus, and grammar fall out of different (V, Γ) choices.
+
+## Model
+
+- Architecture: Transformer decoder (custom, mechanism-aligned)
+- Parameters: 10.5M
+- Training tokens: ~200K (derivation traces)
+- Training epochs: 5
+
+## Benchmark: DTA (Derivation Trace Accuracy)
+
+The correct benchmark for this model is not BLiMP or MMLU.
+It is DTA — how accurately does the model predict forced boundary conditions
+on novel carriers?
+
+See: `ApplePiesFromScratch/dta-benchmark`
+
+| Model | DTA-Overall | DTA-Closure | DTA-FixedPts | DTA-Cycle |
+|-------|-------------|-------------|--------------|-----------|
+| MechanismBase (10M) | TBD | TBD | TBD | TBD |
+| Random baseline | 25% | 50% | 25% | 25% |
+| Engine (oracle) | 100% | 100% | 100% | 100% |
+
+## Usage
+
+```python
+# The model requires the pl/ library and engine.py from the repo
+# Clone: github.com/ApplePiesFromScratch/propagation-logic
+
+from model import MechanismBase, SmallConfig
+from tokenizers import Tokenizer
+import torch
+
+config = SmallConfig()
+model = MechanismBase(config)
+# Load weights from Hub (see full usage in repo)
+
+tokenizer = Tokenizer.from_file("mechanism_tokenizer/tokenizer.json")
+
+# Give the model a partial derivation trace
+partial = """DOMAIN: color_domain
+CARRIER: ['red', 'green', 'blue']
+GRADIENTS: ['complement', 'id']
+THETA: 1.0
+---
+"""
+
+ids = torch.tensor(tokenizer.encode(partial).ids).unsqueeze(0)
+output = model.generate(ids, max_new_tokens=200, temperature=0.3)
+print(tokenizer.decode(output[0].tolist()))
+```
+
+## Training
+
+```
+python generate_data.py    # generates derivation trace corpus
+python tokenizer_train.py  # BPE tokenizer on corpus
+python train.py            # SmallConfig, ~30 min on RTX 4060 Ti
+```
+
+## Repository
+
+GitHub: [ApplePiesFromScratch/propagation-logic](https://github.com/ApplePiesFromScratch/propagation-logic)

mechanism_base_v1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7212a180679ab3d408d6c23d3904b9b9d3b06c9a34b4287c6e788910738daafb
+oid sha256:e326fabf30f2474aa3e5af5eb89c0e300cefdbaa7c6cefb87243f894c4545dda
 size 42349189