qwen2-mic

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

• Contract manifest
• Transformers usage
• Transformers expectations
• Fixture provenance

Contract

qwen2 model integration contract

Public Transformers usage validated on a slim fixture on an NVIDIA A10G, with default, eager and bfloat16 variants and fresh-process baseline recording. This prototype declares fixture-only targets; full checkpoint coverage is separate.

Regeneration evidence.

Slim model

Slim fixture: mic-qwen2-slim

A 9.8M-parameter fixture for the qwen2 MIC, covering Qwen/Qwen2-0.5B. It is a test artifact, not a usable model, tagged test-artifact to keep it out of candidate-model search.

• Fixture key df1b08640fd5. It covers every checkpoint whose processor files are byte-identical and whose config matches after shrinking.
• Same architecture, smaller config: Qwen2ForCausalLM, shrunk by the generic rule in sweep.py (num_hidden_layers 2).
• Processor files copied byte for byte from Qwen/Qwen2-0.5B: LICENSE, generation_config.json, merges.txt, tokenizer.json, tokenizer_config.json, vocab.json.
• 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: attention_mask:all_ones, 0.289), so no training was needed.
• Same tensor names as the real checkpoint. The real checkpoint has every parameter, and so does the fixture.
• Saved in bfloat16, the dtype Qwen/Qwen2-0.5B 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 mic-qwen2-slim.