google/paxml issues and pull requests

#100 - docs: update README.md

Pull Request - State: open - Opened by eltociear 6 months ago

#99 - Bump tensorflow from 2.9.3 to 2.12.1 in /paxml/pip_package in the pip group across 1 directory

Pull Request - State: open - Opened by dependabot[bot] 6 months ago
Labels: dependencies

#98 - Add cuda profiler hook

Pull Request - State: open - Opened by shawnwang18 7 months ago

#97 - Bump the pip group in /paxml/pip_package with 2 updates

Pull Request - State: closed - Opened by dependabot[bot] 7 months ago - 1 comment
Labels: dependencies

#96 - LoRA support in transformers

Issue - State: open - Opened by tanmayshishodia 7 months ago - 1 comment

#95 - Error running tutorial notebook

Issue - State: open - Opened by ajikmr 8 months ago

#94 - Bump the pip group across 1 directory with 3 updates

Pull Request - State: closed - Opened by dependabot[bot] 8 months ago - 1 comment
Labels: dependencies

#93 - Add flag to use dpa in grok

Pull Request - State: open - Opened by hx89 8 months ago

#92 - Bump the pip group across 1 directory with 2 updates

Pull Request - State: closed - Opened by dependabot[bot] 8 months ago - 1 comment
Labels: dependencies

#91 - Support fractional PERCORE_BATCH_SIZE in synthetic dataset

Pull Request - State: closed - Opened by hx89 8 months ago
Labels: pull ready

#90 - Update README.md

Pull Request - State: open - Opened by eltociear 8 months ago

#89 - lingvo issue while installing paxml in vscode

Issue - State: open - Opened by juneedpk 9 months ago

#88 - Bump the pip group across 1 directory with 2 updates

Pull Request - State: closed - Opened by dependabot[bot] 9 months ago - 1 comment
Labels: dependencies

#87 - Add pipeline parallelism for Grok

Pull Request - State: closed - Opened by hx89 9 months ago - 1 comment
Labels: pull ready

#86 - update grok model param

Pull Request - State: closed - Opened by hx89 9 months ago - 1 comment
Labels: pull ready

#85 - Bump tensorflow from 2.9.3 to 2.11.1 in /paxml/pip_package in the pip group across 1 directory

Pull Request - State: closed - Opened by dependabot[bot] 9 months ago - 1 comment
Labels: dependencies

#84 - Bump the pip group across 1 directory with 3 updates

Pull Request - State: closed - Opened by dependabot[bot] 9 months ago - 1 comment
Labels: dependencies

#83 - [NVIDIA] Add LLaMA SFT and LoRA PEFT support

Pull Request - State: closed - Opened by ashors1 9 months ago - 1 comment
Labels: pull ready

#82 - [NVIDIA] Add MoE Configs

Pull Request - State: closed - Opened by ashors1 9 months ago
Labels: pull ready

#81 - Revert "[NVIDIA ] Remove references to deprecated XLA flags."

Pull Request - State: closed - Opened by laurentes 9 months ago

#80 - [NVIDIA ] Remove references to deprecated XLA flags.

Pull Request - State: closed - Opened by Tixxx 9 months ago

#79 - Added new config to nvidia.py

Pull Request - State: closed - Opened by abhinavgoel95 10 months ago
Labels: pull ready

#78 - adding new configs to nvidia.py

Pull Request - State: closed - Opened by abhinavgoel95 10 months ago - 2 comments

#77 - DEADLINE_EXCEEDED on 1024 GPUs.

Issue - State: open - Opened by mhugues 10 months ago

#76 - Bump idna from 3.6 to 3.7 in /paxml/pip_package

Pull Request - State: closed - Opened by dependabot[bot] 10 months ago - 1 comment
Labels: dependencies

#75 - Adding support for expert parallelism

Pull Request - State: closed - Opened by abhinavgoel95 10 months ago
Labels: pull ready

#74 - Bump pillow from 10.2.0 to 10.3.0 in /paxml/pip_package

Pull Request - State: closed - Opened by dependabot[bot] 10 months ago - 1 comment
Labels: dependencies

#73 - [NVIDIA] Add config option to use cudnn flash attention

Pull Request - State: closed - Opened by kaixih 11 months ago - 2 comments
Labels: pull ready

#72 - Set offload checkpoint policy

Pull Request - State: open - Opened by jaro-sevcik 11 months ago

#71 - Jax + tpu and AQT int8 train model loss is abnormal

Issue - State: open - Opened by Lisennlp 11 months ago

#70 - [NVIDIA] Configure steps/sec interval to compute steps/sec only when writing a summary

Pull Request - State: closed - Opened by ashors1 12 months ago
Labels: pull ready

#69 - Make `compute_steps_per_sec_interval_steps` configurable

Pull Request - State: closed - Opened by ashors1 12 months ago
Labels: pull ready

#68 - [NVIDIA] Add LLaMA configs and scripts

Pull Request - State: closed - Opened by ashors1 12 months ago - 4 comments
Labels: pull ready

#67 - Set eval_loop_num_batches=-1 for lambada eval

Pull Request - State: closed - Opened by ashors1 about 1 year ago
Labels: pull ready

#66 - Use bfloat16 for eval

Issue - State: open - Opened by tbaker2 about 1 year ago - 1 comment

#65 - [Question] Very low MFU(30%~35%) when train bf16 Llama2 and GPT model with single SXM4 A100 machine.

Issue - State: open - Opened by MoFHeka about 1 year ago

#64 - [Feature Request] Need ZeRo-1/2 to cooperate with PP+TP+DP. Which may more faster than FSDP sometimes.

Issue - State: open - Opened by MoFHeka about 1 year ago

#63 - Bump notebook from 7.0.6 to 7.0.7 in /paxml/pip_package

Pull Request - State: closed - Opened by dependabot[bot] about 1 year ago - 1 comment
Labels: dependencies

#62 - Bump jupyterlab from 4.0.10 to 4.0.11 in /paxml/pip_package

Pull Request - State: closed - Opened by dependabot[bot] about 1 year ago - 1 comment
Labels: dependencies

#61 - [NVIDIA] Add synthetic configs for benchmarking

Pull Request - State: closed - Opened by ashors1 about 1 year ago
Labels: pull ready

#60 - Bump jupyter-lsp from 2.2.1 to 2.2.2 in /paxml/pip_package

Pull Request - State: closed - Opened by dependabot[bot] about 1 year ago - 1 comment
Labels: dependencies

#59 - Bump jinja2 from 3.1.2 to 3.1.3 in /paxml/pip_package

Pull Request - State: closed - Opened by dependabot[bot] about 1 year ago - 1 comment
Labels: dependencies

#59 - Bump jinja2 from 3.1.2 to 3.1.3 in /paxml/pip_package

Pull Request - State: open - Opened by dependabot[bot] about 1 year ago
Labels: dependencies

#58 - Remaining tutorials

Issue - State: open - Opened by rahulbatra85 about 1 year ago

#57 - Bump jupyter-server from 2.11.1 to 2.11.2 in /paxml/pip_package

Pull Request - State: closed - Opened by dependabot[bot] about 1 year ago - 1 comment
Labels: dependencies

#56 - [NVIDIA] Add Llama2 configs

Pull Request - State: closed - Opened by ashors1 about 1 year ago
Labels: pull ready

#55 - [NVIDIA] Simplify the unit test for overwrite_with_gradient

Pull Request - State: closed - Opened by kaixih about 1 year ago
Labels: pull ready

#54 - Convert string dtype to jnp dtype during evaluation

Pull Request - State: closed - Opened by ashors1 about 1 year ago
Labels: pull ready

#53 - Redirect to NVIDIA Rosetta from GPU section of README

Pull Request - State: closed - Opened by ashors1 over 1 year ago
Labels: pull ready

#52 - [NVIDIA] Update 5B config and scripts

Pull Request - State: closed - Opened by ashors1 over 1 year ago
Labels: pull ready

#51 - Bump urllib3 from 2.0.6 to 2.0.7 in /paxml/pip_package

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago - 1 comment
Labels: dependencies

#50 - Bump urllib3 from 1.26.16 to 1.26.17 in /paxml/pip_package

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago - 1 comment
Labels: dependencies

#49 - [NVIDIA] Support new config option `USE_FP8`

Pull Request - State: closed - Opened by kaixih over 1 year ago - 1 comment
Labels: pull ready

#48 - [NVIDIA] New collection for variables 'overwrite_with_gradient'

Pull Request - State: closed - Opened by kaixih over 1 year ago
Labels: pull ready

#47 - [NVIDIA] Add 27B MoE config

Pull Request - State: closed - Opened by ashors1 over 1 year ago
Labels: pull ready

#46 - Add Transformer Engine support to Paxml

Pull Request - State: open - Opened by ashors1 over 1 year ago

#45 - Update GPU Scripts and Configs

Pull Request - State: closed - Opened by ashors1 over 1 year ago
Labels: pull ready

#44 - Support FP8 params updating for NVIDIA Hopper GPUs

Pull Request - State: closed - Opened by kaixih over 1 year ago - 3 comments

#43 - Use CPU-only version of TensorFlow

Pull Request - State: closed - Opened by andportnoy over 1 year ago - 1 comment

#42 - Bump tornado from 6.3.2 to 6.3.3 in /paxml/pip_package

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago - 1 comment
Labels: dependencies

#41 - [NVIDIA] Update scripts to improve configurability

Pull Request - State: closed - Opened by ashors1 over 1 year ago
Labels: pull ready

#40 - support for fractional per core batch size

Pull Request - State: open - Opened by abhinavgoel95 over 1 year ago - 1 comment

#39 - Update README.md: fix missing -r argument to pip

Pull Request - State: open - Opened by joker-eph over 1 year ago - 1 comment
Labels: pull ready

#38 - fix broken link in readme

Pull Request - State: closed - Opened by sadikneipp over 1 year ago
Labels: pull ready

#37 - How to continue training from a checkpoint?

Issue - State: open - Opened by lkm2835 over 1 year ago

#36 - [NVIDIA] Fix Default Container in Example Submit File

Pull Request - State: closed - Opened by ashors1 over 1 year ago
Labels: pull ready

#35 - Int8 checkpoint

Issue - State: open - Opened by wx-x over 1 year ago

#34 - [NVIDIA] Update GPU Documentation, Configs, and Paths

Pull Request - State: closed - Opened by ashors1 over 1 year ago - 1 comment
Labels: pull ready

#33 - Add multislice configs and documentation

Pull Request - State: closed - Opened by michelle-yooh over 1 year ago - 1 comment

#32 - Bump transformers from 4.29.2 to 4.30.0 in /paxml/pip_package

Pull Request - State: open - Opened by dependabot[bot] over 1 year ago
Labels: dependencies

#31 - Bump transformers from 4.27.4 to 4.30.0 in /paxml/contrib/gpu/scripts_gpu

Pull Request - State: open - Opened by dependabot[bot] over 1 year ago
Labels: dependencies

#30 - ARM64 Build

Issue - State: open - Opened by joker-eph over 1 year ago - 4 comments

#29 - Bump tornado from 6.2 to 6.3.2 in /paxml/contrib/gpu/scripts_gpu

Pull Request - State: open - Opened by dependabot[bot] over 1 year ago
Labels: dependencies

#28 - Bump requests from 2.30.0 to 2.31.0 in /paxml/pip_package

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago - 1 comment
Labels: dependencies

#27 - Bump requests from 2.28.2 to 2.31.0 in /paxml/contrib/gpu/scripts_gpu

Pull Request - State: open - Opened by dependabot[bot] over 1 year ago
Labels: dependencies

#26 - Anisha lg pax

Pull Request - State: open - Opened by A9isha over 1 year ago

#25 - Installing paxml from source failed due to dependency problem

Issue - State: closed - Opened by yhtang over 1 year ago - 5 comments

#24 - Pin JAX version in Dockerfile

Pull Request - State: closed - Opened by ashors1 over 1 year ago
Labels: pull ready

#23 - Correct referenced commit in gpu README

Pull Request - State: closed - Opened by ashors1 over 1 year ago
Labels: pull ready

#22 - Yooh multislice

Pull Request - State: closed - Opened by michelle-yooh almost 2 years ago

#21 - Enable Latency Hiding Scheduler in Dockerfile

Pull Request - State: closed - Opened by ashors1 almost 2 years ago
Labels: pull ready

#20 - Bump tensorflow from 2.9.3 to 2.11.1 in /paxml/contrib/gpu/scripts_gpu

Pull Request - State: closed - Opened by dependabot[bot] almost 2 years ago - 1 comment
Labels: dependencies

#19 - Update nvidia.py to add more configs

Pull Request - State: closed - Opened by abhinavgoel95 almost 2 years ago - 2 comments
Labels: pull ready

#17 - Unexpected Overheads with Activation Checkpointing with Pipeline Parallelism

Issue - State: open - Opened by abhinavgoel95 almost 2 years ago - 1 comment

#16 - Update GPU Dockerfile and Configs

Pull Request - State: closed - Opened by ashors1 almost 2 years ago
Labels: pull ready

#15 - Bump tensorflow from 2.9.3 to 2.11.1 in /paxml/pip_package

Pull Request - State: closed - Opened by dependabot[bot] almost 2 years ago - 1 comment
Labels: pull ready, dependencies

#14 - Make num_train_steps configurable in gpu configs

Pull Request - State: closed - Opened by ashors1 almost 2 years ago
Labels: pull ready

#13 - Update Dockerfile and Build Path

Pull Request - State: closed - Opened by ashors1 almost 2 years ago
Labels: pull ready

#12 - Update Logging for Gradient Accumulation

Pull Request - State: closed - Opened by ashors1 almost 2 years ago
Labels: pull ready

#11 - Error running Common Crawl example

Issue - State: closed - Opened by RobertLiJN almost 2 years ago - 2 comments

#10 - Add GPU scripts and dependencies

Pull Request - State: closed - Opened by ashors1 almost 2 years ago - 3 comments
Labels: pull ready

#9 - Perform gradient clipping on global batch when using gradient accumulation

Pull Request - State: open - Opened by ashors1 almost 2 years ago - 3 comments
Labels: pull ready

#8 - Update Metrics and Flags for GPU

Pull Request - State: closed - Opened by ashors1 almost 2 years ago - 1 comment
Labels: pull ready

#7 - Convert string to jnp dtype

Pull Request - State: closed - Opened by ashors1 almost 2 years ago
Labels: pull ready

#6 - update docstring

Pull Request - State: closed - Opened by ashors1 about 2 years ago - 2 comments
Labels: pull ready

#5 - Pipeline Parallelism: USE_REPEATED_LAYERS bug

Issue - State: closed - Opened by abhinavgoel95 over 2 years ago - 3 comments

#4 - Pipeline Parallelism: F external/org_tensorflow/tensorflow/compiler/xla/array.h:446] Check failed: n < sizes_size Fatal Python error: Aborted

Issue - State: closed - Opened by abhinavgoel95 over 2 years ago - 3 comments

GitHub / google/paxml issues and pull requests