File size: 2,490 Bytes
5ecff7c d510ef0 5ecff7c 565fb17 5ecff7c d510ef0 5ecff7c 565fb17 d510ef0 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 | ---
tags:
- mic
- test-artifact
base_model: google-bert/bert-base-uncased
---
# bert-mic
Slim model weights, executable usage and framework expectations in one repository.
- [Contract manifest](integration.json)
- [Transformers usage](integrations/transformers/usage.md)
- [Transformers expectations](integrations/transformers/expectations)
- [Fixture provenance](build.json)
## Contract
### bert 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](https://github.com/huggingface/transformers-mic/actions/runs/38252430459).
## Slim model
### Slim fixture: mic-bert-slim
A 2.1M-parameter fixture for the [bert MIC](https://huggingface.co/huggingface-mic/bert-mic), covering [`google-bert/bert-base-uncased`](https://huggingface.co/google-bert/bert-base-uncased). It is a test artifact, not a usable model, tagged `test-artifact` to keep it out of candidate-model search.
- **Fixture key `7ce74647df48`.** It covers every checkpoint whose processor files are byte-identical and whose config matches after shrinking.
- **Same architecture, smaller config:** `BertForMaskedLM`, shrunk by the generic rule in [`sweep.py`](https://github.com/huggingface/transformers-mic/blob/main/experiments/sweep.py) (num_hidden_layers 2).
- **Processor files copied byte for byte** from `google-bert/bert-base-uncased`: `LICENSE`, `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: `attention_mask:all_ones`, 0.00278), 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 float32**, the dtype `google-bert/bert-base-uncased` loads in by default.
- **Finite in float32, bfloat16, and float16** (forward, loss, gradients).
Built by [`build_fixtures.py`](https://github.com/huggingface/transformers-mic/blob/main/example/build_fixtures.py); [`build.json`](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-bert-slim`.
|