transformers/docs/source/en
Abhi Venigalla 005b957fb8
Add DBRX Model (#29921)
* wip

* fix __init__.py

* add docs

* Apply suggestions from code review

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* address comments 1

* work on make fixup

* pass configs down

* add sdpa attention

* remove DbrxBlock

* add to configuration_auto

* docstring now passes formatting test

* fix style

* update READMEs

* add dbrx to modeling_auto

* make fix-copies generated this

* add DBRX_PRETRAINED_CONFIG_ARCHIVE_MAP

* config docstring passes formatting test

* rename moe_loss_weight to router_aux_loss_coef

* add to flash-attn documentation

* fix model-path in tests

* Explicitly make `"suli"` the default `ffn_act_fn`

Co-authored-by: Wing Lian <wing.lian@gmail.com>

* default to using router_aux_loss_coef over ffn_config[moe_loss_weight]

* fix _flash_attn_uses_top_left_mask and is_causal

* fix tests path

* don't use token type IDs

* follow Llama and remove token_type_ids from test

* init ConfigTester differently so tests pass

* remove multiple choice test

* remove question + answer test

* remove sequence classification test

* remove token classification test

* copy Llama tests and remove token_type_ids from test inputs

* do not test pruning or headmasking; style code

* add _tied_weights_keys parameter to pass test

* add type hints

* fix type check

* update config tester

* remove masked_lm test

* remove encoder tests

* initialize DbrxModelTester with correct params

* style

* torch_dtype does not rely on torch

* run make fixup, fix-copies

* use https://huggingface.co/v2ray/dbrx-base-fixed/blob/main/modeling_dbrx.py

* add copyright info

* fix imports and DbrxRotaryEmbedding

* update DbrxModel docstring

* use copies

* change model path in docstring

* use config in DbrxFFN

* fix flashattention2, sdpaattention

* input config to DbrXAttention, DbrxNormAttentionNorm

* more fixes

* fix

* fix again!

* add informative comment

* fix ruff?

* remove print statement + style

* change doc-test

* fix doc-test

* fix docstring

* delete commented out text

* make defaults match dbrx-instruct

* replace `router_aux_loss_coef` with `moe_loss_weight`

* is_decoder=True

* remove is_decoder from configtester

* implement sdpa properly

* make is_decoder pass tests

* start on the GenerationTesterMixin tests

* add dbrx to sdpa documentation

* skip weight typing test

* style

* initialize smaller model

Co-authored-by: Matt <Rocketknight1@users.noreply.github.com>

* Add DBRX to toctree

* skip test_new_cache_format

* make config defaults smaller again

* add pad_token_id

* remove pad_token_id from config

* Remove all references to DBRX_PRETRAINED_CONFIG_ARCHIVE_MAP

* Update src/transformers/models/dbrx/__init__.py

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Update src/transformers/models/dbrx/modeling_dbrx.py

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Update docs/source/en/model_doc/dbrx.md

Co-authored-by: Matt <Rocketknight1@users.noreply.github.com>

* Update src/transformers/models/dbrx/configuration_dbrx.py

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Update docs/source/en/model_doc/dbrx.md

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* fix typo

* Apply suggestions from code review

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* update docs, fix configuration_auto.py

* address pr comments

* remove is_decoder flag

* slice

* fix requires grad

* remove grad

* disconnect differently

* remove grad

* enable grads

* patch

* detach expert

* nissan al ghaib

* Update modeling_dbrx.py

* Update src/transformers/models/dbrx/modeling_dbrx.py

Co-authored-by: Matt <Rocketknight1@users.noreply.github.com>

* replace "Gemma" with "Dbrx"

* remove # type: ignore

* don't hardcode vocab_size

* remove ToDo

* Re-add removed idefics2 line

* Update test to use tiny-random!

* Remove TODO

* Remove one more case of loading the entire dbrx-instruct in the tests

* Update src/transformers/models/dbrx/modeling_dbrx.py

Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

* address some comments

* small model

* add dbrx to tokenization_auto

* More docstrings with add_start_docstrings

* Dbrx for now

* add PipelineTesterMixin

* Update src/transformers/models/dbrx/configuration_dbrx.py

Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

* remove flash-attn2 import error

* fix docstring

Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

* add useage example

* put on one line

Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

* fix ffn_act_fn

Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

* change "dbrx" to "DBRX" for display purposes.

* fix __init__.py?

* fix __init__.py

* fix README

* return the aux_loss

* remove extra spaces

* fix configuration_auto.py

* fix format in tokenization_auto

* remove new line

* add more useage examples

---------

Co-authored-by: Abhi Venigalla <abhi.venigalla@databricks.com>
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
Co-authored-by: Eitan Turok <eitan.turok@databricks.com>
Co-authored-by: Eitan Turok <150733043+eitanturok@users.noreply.github.com>
Co-authored-by: Wing Lian <wing.lian@gmail.com>
Co-authored-by: Eitan Turok <eitanturok@gmail.com>
Co-authored-by: Matt <Rocketknight1@users.noreply.github.com>
Co-authored-by: Matt <rocketknight1@gmail.com>
Co-authored-by: Your Name <you@example.com>
Co-authored-by: Mihir Patel <mihir.v.patel7@gmail.com>
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
2024-04-18 15:18:52 +02:00
..
internal v4.39 deprecations 🧼 (#29492) 2024-03-07 10:44:43 +00:00
main_classes [docs] Big model loading (#29920) 2024-04-01 18:47:32 -07:00
model_doc Add DBRX Model (#29921) 2024-04-18 15:18:52 +02:00
tasks Add DBRX Model (#29921) 2024-04-18 15:18:52 +02:00
_config.py [#29174] ImportError Fix: Trainer with PyTorch requires accelerate>=0.20.1 Fix (#29888) 2024-04-08 14:21:16 +01:00
_redirects.yml Extended semantic segmentation to image segmentation (#27039) 2023-11-23 15:58:21 +00:00
_toctree.yml Add DBRX Model (#29921) 2024-04-18 15:18:52 +02:00
accelerate.md Fix typos (#25936) 2023-09-04 11:15:12 +01:00
add_new_model.md [docs] Indent ordered list in add_new_model.md (#29796) 2024-03-26 12:03:39 +00:00
add_new_pipeline.md add `push_to_hub` to pipeline (#29172) 2024-04-16 15:34:04 +01:00
add_tensorflow_model.md [Docs] Make an ordered list prettier in add_tensorflow_model.md (#29949) 2024-04-02 12:37:56 +01:00
attention.md [Docs] Fix broken links and syntax issues (#28918) 2024-02-08 14:13:35 -08:00
autoclass_tutorial.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
benchmarks.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
bertology.md Migrate doc files to Markdown. (#24376) 2023-06-20 18:07:47 -04:00
big_models.md [docs] Big model loading (#29920) 2024-04-01 18:47:32 -07:00
chat_templating.md [docs] Remove broken ChatML format link from chat_templating.md (#29643) 2024-03-13 13:04:51 -07:00
community.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
contributing.md Enable doc in Spanish (#16518) 2022-04-04 10:25:46 -04:00
create_a_model.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
custom_models.md [Docs] Add language identifiers to fenced code blocks (#28955) 2024-02-12 10:48:31 -08:00
custom_tools.md [docs] Remove redundant `-` and `the` from custom_tools.md (#29767) 2024-03-21 10:56:40 +00:00
debugging.md [Docs] Fix spelling and grammar mistakes (#28825) 2024-02-02 08:45:00 +01:00
deepspeed.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
fast_tokenizers.md Migrate doc files to Markdown. (#24376) 2023-06-20 18:07:47 -04:00
fsdp.md [docs] Trainer docs (#28145) 2023-12-20 10:37:23 -08:00
generation_strategies.md Fix doctest more (for `docs/source/en`) (#30247) 2024-04-15 14:10:59 +02:00
glossary.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
hf_quantizer.md [CI] Quantization workflow (#29046) 2024-02-28 10:09:25 -05:00
hpo_train.md Remove-auth-token (#27060) 2023-11-13 14:20:54 +01:00
index.md Add DBRX Model (#29921) 2024-04-18 15:18:52 +02:00
installation.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
llm_tutorial.md Generate: All logits processors are documented and have examples (#27796) 2023-12-07 15:11:35 +00:00
llm_tutorial_optimization.md F.scaled_dot_product_attention support (#26572) 2023-12-09 05:38:14 +09:00
model_memory_anatomy.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
model_sharing.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
model_summary.md model_summary.md - Restore link to Harvard's Annotated Transformer. (#29702) 2024-03-23 18:29:39 -07:00
multilingual.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
notebooks.md Enable doc in Spanish (#16518) 2022-04-04 10:25:46 -04:00
pad_truncation.md [Doc] Spanish translation of pad_truncation.md (#27890) 2023-12-08 10:32:18 -08:00
peft.md [`Peft`] `modules_to_save` support for peft integration (#27466) 2023-11-14 10:32:57 +01:00
perf_hardware.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
perf_infer_cpu.md [Docs] Fix spelling and grammar mistakes (#28825) 2024-02-02 08:45:00 +01:00
perf_infer_gpu_one.md Add DBRX Model (#29921) 2024-04-18 15:18:52 +02:00
perf_torch_compile.md Fix rendering for `torch.compile()` docs (#25432) 2023-08-10 13:25:00 +02:00
perf_train_cpu.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
perf_train_cpu_many.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
perf_train_gpu_many.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
perf_train_gpu_one.md Fix minor typo: softare => software (#29602) 2024-03-12 10:39:56 +00:00
perf_train_special.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
perf_train_tpu_tf.md Migrate doc files to Markdown. (#24376) 2023-06-20 18:07:47 -04:00
performance.md [docs] Update CPU/GPU inference docs (#26881) 2023-10-31 09:44:51 -07:00
perplexity.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
philosophy.md [docs] fixed links with 404 (#27327) 2023-11-06 19:45:03 +00:00
pipeline_tutorial.md More fixes for doctest (#30265) 2024-04-16 11:58:55 +02:00
pipeline_webserver.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
pr_checks.md [Docs] Fix spelling and grammar mistakes (#28825) 2024-02-02 08:45:00 +01:00
preprocessing.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
quantization.md [Quantization] Quanto quantizer (#29023) 2024-03-15 11:51:29 -04:00
quicktour.md [#29174] ImportError Fix: Trainer with PyTorch requires accelerate>=0.20.1 Fix (#29888) 2024-04-08 14:21:16 +01:00
run_scripts.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
sagemaker.md [docs] fixed links with 404 (#27327) 2023-11-06 19:45:03 +00:00
serialization.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
task_summary.md More fixes for doctest (#30265) 2024-04-16 11:58:55 +02:00
tasks_explained.md [docs] Spanish translation of tasks_explained.md (#29224) 2024-02-26 08:18:15 -08:00
testing.md [doc] fix some typos and add `xpu` to the testing documentation (#29894) 2024-03-28 09:42:49 +00:00
tf_xla.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
tflite.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
tokenizer_summary.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
torchscript.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
trainer.md FEAT / Optim: Add GaLore optimizer (#29588) 2024-03-19 11:40:23 +01:00
training.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00
transformers_agents.md [doc] Always call it Agents for consistency (#25958) 2023-09-05 12:27:20 +01:00
troubleshooting.md Update all references to canonical models (#29001) 2024-02-16 08:16:58 +01:00