layoutlm-base-uncased-mic

Slim model weights, executable usage and framework expectations in one repository.

• Contract manifest
• Transformers usage
• Transformers expectations
• Fixture provenance
• How the fixture was shrunk

Contract

LayoutLM masked language modeling

Runnable usage and the framework expectations directory are linked above.

Slim model

Slim fixture: layoutlm-base-uncased-mic

A 2.4M-parameter fixture for the LayoutLM MIC, covering microsoft/layoutlm-base-uncased. It is a test artifact, not a usable model, tagged test-artifact to keep it out of candidate-model search.

• Fixture key e6e5f39b6b0e. It covers every checkpoint whose processor files are byte-identical and whose config matches after shrinking.
• Same architecture, smaller config: LayoutLMForMaskedLM, shrunk by the generic rule in sweep.py (num_hidden_layers 2).
• Processor files copied byte for byte from microsoft/layoutlm-base-uncased: special_tokens_map.json, tokenizer.json, tokenizer_config.json, vocab.txt.
• Random weights (seed 17). They passed the input-sensitivity gate: every perturbation of the contract's inputs moved the logits by at least 0.001 (weakest: input_ids:roll_batch, 0.356), so no training was needed.
• Same tensor names as the real checkpoint. The real checkpoint lacks cls.predictions.bias, cls.predictions.decoder.bias, cls.predictions.transform.LayerNorm.bias, cls.predictions.transform.LayerNorm.weight, cls.predictions.transform.dense.bias, cls.predictions.transform.dense.weight, so the fixture omits it too, and loading initializes it as it does for users.
• Saved in float32, the dtype microsoft/layoutlm-base-uncased loads in by default.
• Finite in float32, bfloat16, and float16 (forward, loss, gradients).

Built by build_fixtures.py; build.json records the pinned sources, gate results, and file hashes. This repository includes the slim weights, usage and framework expectations; rebuild the other files with python example/build_fixtures.py --fixture layoutlm-base-uncased-mic.