9 Commits

Author	SHA1	Message	Date
Yih-Dar	05de764e9c	Aurevoir PyTorch 1 (#35358) ... * fix * fix * fix --------- Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>	2024-12-20 14:36:31 +01:00
Anton Vlasjuk	5a2aedca1e	[Mamba2] Fix caching, slow path, and multi-gpu (#35154) ... * fixup mamba2 - caching and several other small fixes * fixup cached forward * correct fix this time * fixup cache - we do not need to extend the attn mask it's handled by generate (gives total ids + mask at each step) * remove unnecessary (un)squeeze * fixup cache position * simplify a few things * [run-slow] mamba2 * multi gpu attempt two * [run-slow] mamba2 * [run-slow] mamba2 * [run-slow] mamba2 * [run-slow] mamba2 * add newer slow path fix * [run-slow] mamba2	2024-12-20 09:27:47 +01:00
Joao Gante	8a734ea2c3	Tests: move generate tests to the right mixin and delete redundant tests (#34464) ... * tmp commit * tmp commit * cull overwrites of deleted tests * typo * more specific docstring * make fixup * parameterize at the top? * correction * more deletions :D * tmp commit * for VLMs too * fix _check_outputs * test nit * make fixup * fix another flaky * test_generate_from_inputs_embeds -- handle missing attention mask	2024-10-30 10:59:08 +00:00
Raushan Turganbay	808d6c50f8	Generation: fix test (#34369) ... * fix test * fix copies	2024-10-29 07:57:10 +01:00
Lake Lee	ec1424c6a3	Update modeling_mamba2.py, fix pad size (#32599) ... * Update modeling_mamba2.py Fix pad_size calculation to ensure it's less than self.chunk_size * [run_slow] mamba2 * [run-slow] mamba2 * [run-slow] Add @require_read_token decorator to failing tests for token propagation * [run_slow] mamba2	2024-09-20 11:40:57 +01:00
Anton Vlasjuk	b50ff5993a	[Mamba2] Move dt calculations to kernel (#33520) ... * use kernel for dt calculations * add small test * [run-slow] mamba2	2024-09-19 17:41:17 +01:00
Pablo Montalvo	4f0246e535	fix tests with main revision and read token (#33560) ... * fix tests with main revision and read token * [run-slow]mamba2 * test previously skipped tests * [run-slow]mamba2 * skip some tests * [run-slow]mamba2 * finalize tests * [run-slow]mamba2	2024-09-19 17:10:22 +02:00
Joao Gante	70d5df6107	Generate: unify LogitsWarper and LogitsProcessor (#32626)	2024-08-16 11:20:41 +01:00
Pablo Montalvo	80b90e7b2f	Add codestral mamba2 (#32080) ... * add new model like * draft cuda forward - mismatched keys (sharding on conv1) * match keys successfully * fix split * get generation/forward running (wrong gens, norm?) * :update * some refactoring * fixes * works up until copy to cache * fix * update * NON WORKING VERSION * version that work? * nit * fix config * fix conversion script * working cuda forward * nit * update * simplifcation * make mamba slow simple work * no einops * todo * fix style * no einops * update fix no einsum * nit * remove einops * bug: scan_output differs strongly * add rms norm option * fix fast + slow generation with and w/o cache ✔️ * draft integration tests * remove a big chunk of the einsum * fix slow, fast generations, without any einsum * fix copies * fix structure * fix up modeling and tests * fix tests * clamping is indeed worse * recover mamba2 cache test * fix copies * no cache position (yet) * fix tf tests * fix matmul for generate * fixup * skip cache tests for now * [run-slow]mamba2 * tune out hidden states for padding * test batched generation * propagate attention mask changes * fix past length * fix integration test * style * address comments * update readme * add mamba2 version check * fix tests * [run-slow]mamba2 * skip edge tests * [run-slow]mamba2 * last fixup * [run-slow]mamba2 * update README --------- Co-authored-by: Arthur Zucker <arthur.zucker@gmail.com>	2024-08-06 16:39:52 +02:00