* uneql rank. * trim. * enable passing in number of heads for each rank. * simplify. * simplify. * cleanup. * fix col parallel. * fix bug with row parallel. * fit out proj. * refac. * fix sharding logic. * refac sharding. * refac. * support multiple of. * make fn reuseable. * fix bug in dimensions. * scaffold. * test uneven heads. * fix test by adding barrier. * refac. * reuse code. * clean up. |
||
|---|---|---|
| .. | ||
| __init__.py | ||
| bert.py | ||
| falcon.py | ||
| gpt_neox.py | ||
| gpt.py | ||
| gptj.py | ||
| llama.py | ||
| opt.py | ||
| vit.py | ||