-
Notifications
You must be signed in to change notification settings - Fork 0
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add Comet integration #1
base: main
Are you sure you want to change the base?
Commits on Jun 11, 2024
-
Configuration menu - View commit details
-
Copy full SHA for 52db752 - Browse repository at this point
Copy the full SHA 52db752View commit details -
Configuration menu - View commit details
-
Copy full SHA for 463aad9 - Browse repository at this point
Copy the full SHA 463aad9View commit details
Commits on Jun 13, 2024
-
Add configuration options for customizing the experiment name, tags a…
…nd others fields
Configuration menu - View commit details
-
Copy full SHA for 804cbcc - Browse repository at this point
Copy the full SHA 804cbccView commit details -
Configuration menu - View commit details
-
Copy full SHA for 108f583 - Browse repository at this point
Copy the full SHA 108f583View commit details -
Configuration menu - View commit details
-
Copy full SHA for 90ef7ae - Browse repository at this point
Copy the full SHA 90ef7aeView commit details -
Configuration menu - View commit details
-
Copy full SHA for 12e064b - Browse repository at this point
Copy the full SHA 12e064bView commit details -
Configuration menu - View commit details
-
Copy full SHA for d0b68d7 - Browse repository at this point
Copy the full SHA d0b68d7View commit details
Commits on Jun 19, 2024
-
fix python version and pytest install (EleutherAI#1234)
* fix python version and pytest install * Update NeoXArgs docs automatically * python3 * Update NeoXArgs docs automatically * pip not pip3 * Update NeoXArgs docs automatically * python3 pip * Update NeoXArgs docs automatically * python3 -m pip * Update NeoXArgs docs automatically * Update NeoXArgs docs automatically * Update NeoXArgs docs automatically * add docker setup to workflow * Update NeoXArgs docs automatically * python setup * Update NeoXArgs docs automatically * python setup v2 * Update NeoXArgs docs automatically * python setup v3 * python setup v3 * Update NeoXArgs docs automatically * python setup v3 * Update NeoXArgs docs automatically * python setup v3 * Update NeoXArgs docs automatically * python setup v3 * Update NeoXArgs docs automatically * python setup v3 * python setup v3 * Update NeoXArgs docs automatically * python setup v3 * Update NeoXArgs docs automatically * python setup v3 * Update NeoXArgs docs automatically * python setup v3 * Update NeoXArgs docs automatically * python setup v3 * Update NeoXArgs docs automatically * python setup v3 * Update NeoXArgs docs automatically * python setup v3 * Update NeoXArgs docs automatically * python setup v3 * Update NeoXArgs docs automatically * python setup v3 * Update NeoXArgs docs automatically * python setup v3 * Update NeoXArgs docs automatically * Update NeoXArgs docs automatically * Add hash back to deep speed version * Update NeoXArgs docs automatically --------- Co-authored-by: github-actions <[email protected]> Co-authored-by: Quentin Anthony <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 2608972 - Browse repository at this point
Copy the full SHA 2608972View commit details
Commits on Jun 25, 2024
-
Add a chat data preprocessing script (EleutherAI#1239)
* Add a chat data preprocessing script * add EOT at end of a chat * update README.md * apply pre-commit --------- Co-authored-by: Quentin Anthony <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 0e5f6db - Browse repository at this point
Copy the full SHA 0e5f6dbView commit details
Commits on Jun 28, 2024
-
Configuration menu - View commit details
-
Copy full SHA for 1cee5b7 - Browse repository at this point
Copy the full SHA 1cee5b7View commit details
Commits on Aug 6, 2024
-
Add hf llama to neox conversion (EleutherAI#1247)
* - Add conversion of HF llama models to NeoX * - Add conversion of HF llama models to NeoX * - minor fix * pre-commit --------- Co-authored-by: Quentin Anthony <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for c1ea2a1 - Browse repository at this point
Copy the full SHA c1ea2a1View commit details
Commits on Aug 15, 2024
-
bugfix: chat turns instead of repeating the conversation in preproces…
…s_data_with_chat_template.py (EleutherAI#1258) * bugfix: chat turns instead of repeating the conversation * pre-commit
Configuration menu - View commit details
-
Copy full SHA for 0ef2c07 - Browse repository at this point
Copy the full SHA 0ef2c07View commit details -
Conversion for CI from self-hosted hardware (EleutherAI#1245)
* changing from self-hosted runners to Github's ubuntu-22.04 runner environment * adding warning about not using 'self-hosted' runner labels and using Github runners instead * updated some guidance in comments for coverity scan CI * moving CPU tests to workflow_dispatch only
Configuration menu - View commit details
-
Copy full SHA for f8c9e68 - Browse repository at this point
Copy the full SHA f8c9e68View commit details
Commits on Aug 23, 2024
-
Megatron-LM style Sequence Parallel (EleutherAI#1257)
* first draft (shape errors occurring) * training works (but poor convergence) * debugging progress: current commit works if we do regular TP via impl-ing AR in rowparallel as RS then AG * Update NeoXArgs docs automatically * push most recent code (updated mark_norms fn, back to 'real' sequence parallel) * Update NeoXArgs docs automatically * Fix LayerNorm all reduce gradient hook * Sum instead of average for LayerNorm gradient all reduce * Update NeoXArgs docs automatically * Update NeoXArgs docs automatically * Fix gather and reduce scatter ops on sequence dimension * Fix sequence parallel with tied weight embeddings * Update NeoXArgs docs automatically * cleanup pass + add MoE arguments.py guard * pre-commit and clean up comments * remove vestigial debug code * remove unused debugging code * remove dummy test config * update fp32_allreduce to handle fp16 ; don't cast to fp32 for gathers * run linter on the rest of the files * Improve performance of sequence parallel gather, scatter, and reduce * Add comment * Update NeoXArgs docs automatically --------- Co-authored-by: github-actions <[email protected]> Co-authored-by: Brandon Yang <[email protected]> Co-authored-by: Quentin Anthony <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 8b43196 - Browse repository at this point
Copy the full SHA 8b43196View commit details
Commits on Aug 24, 2024
-
Add new cites (EleutherAI#1255)
* Update README.md I added new models that have come out trained with the GPT-NeoX library. The library itself is sufficiently well-used that simply listing all citing papers is rapidly becoming non-viable. I'm currently leaning towards providing a curated list of "exciting" papers? I haven't looked at other libraries to see what they do yet. * Update NeoXArgs docs automatically --------- Co-authored-by: github-actions <[email protected]> Co-authored-by: Quentin Anthony <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for e7c0182 - Browse repository at this point
Copy the full SHA e7c0182View commit details
Commits on Aug 27, 2024
-
mamba fixes and cleaning (EleutherAI#1262)
* mamba fixes and cleaning * space * revert assertion change for now --------- Co-authored-by: Jacob Hatef <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 591563d - Browse repository at this point
Copy the full SHA 591563dView commit details -
SFT improvements (labeling fixes, different packing implementations) (E…
…leutherAI#1240) * - add different packing impl (Unpacked, packing until overflow) - fix labels to also have valid/test implementations - fix label masking in _get_batch to also include anything from get_ltor_masks_and_position_ids * Update arguments.py to use train_label_data_paths instead of label_data_paths * - fix precommit
Configuration menu - View commit details
-
Copy full SHA for c786367 - Browse repository at this point
Copy the full SHA c786367View commit details
Commits on Sep 3, 2024
-
Configuration menu - View commit details
-
Copy full SHA for 6a2053b - Browse repository at this point
Copy the full SHA 6a2053bView commit details
Commits on Sep 5, 2024
-
Configuration menu - View commit details
-
Copy full SHA for 7548a8b - Browse repository at this point
Copy the full SHA 7548a8bView commit details
Commits on Sep 7, 2024
-
Add
intermediate_size
to GPT-NeoX models (EleutherAI#1212)* Update transformer.py -> Add `intermediate_size` * add support for rwkv and mamba and add todos about swiglu * refactor activations and mlps * change llama config to swiglu * fixes gelu fusion * pre-commit run * add assert message to mamba linear * Update 1-3B.yml revert accidental change * Update 1-3B.yml * fixes various issues * add back swiglu check --------- Co-authored-by: jahatef <[email protected]> Co-authored-by: Quentin Anthony <[email protected]> Co-authored-by: Jacob Hatef <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 0d4bdb9 - Browse repository at this point
Copy the full SHA 0d4bdb9View commit details
Commits on Sep 8, 2024
-
Configuration menu - View commit details
-
Copy full SHA for ec82c05 - Browse repository at this point
Copy the full SHA ec82c05View commit details -
Add DPO training (EleutherAI#1242)
* Add a chat data preprocessing script * add EOT at end of a chat * - add different packing impl (Unpacked, packing until overflow) - fix labels to also have valid/test implementations - fix label masking in _get_batch to also include anything from get_ltor_masks_and_position_ids * update README.md * - Add metrics to forward step to add DPO specific metrics that are useful (accuracy, etc) - Add reference model setup for DPO - Add pairwise dataset for positive/negative pairs - Add DPO loss * Update arguments.py to use train_label_data_paths instead of label_data_paths * - Bugfixes from upstreaming.... * - add precompute logprobs... * - Finishing up precompute logprobs... * - update readme for DPO... * fix varname * Fix pipeline parallelism and incorrect neox_args name * apply precommit --------- Co-authored-by: Quentin Anthony <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 77e8158 - Browse repository at this point
Copy the full SHA 77e8158View commit details
Commits on Sep 9, 2024
-
LayerNorm Refactor (EleutherAI#1269)
* Add TE skeleton * Update NeoXArgs docs automatically * added option for te version of norms * import TERMSNorm * add te norm options to norm arg * add TE objects in weight decay function * reformat * add TERMSNorm and TELayerNorm * Update NeoXArgs docs automatically * - add Fused RMS Norm from apex * - make it consistent with how layernorm looks * Merged transformer engine and apex fused layernorm branches * Added assertion if TE is used * Removed unnecessary transformer-engine import * Changed importerror text for TE * Added requirements/requirements-transformerengine.txt * Add TE skeleton * Update NeoXArgs docs automatically * added option for te version of norms * import TERMSNorm * add te norm options to norm arg * add TE objects in weight decay function * reformat * add TERMSNorm and TELayerNorm * Update NeoXArgs docs automatically * - add Fused RMS Norm from apex * - make it consistent with how layernorm looks * Merged transformer engine and apex fused layernorm branches * Added assertion if TE is used * Removed unnecessary transformer-engine import * Changed importerror text for TE * Added requirements/requirements-transformerengine.txt * update comments * precommit --------- Co-authored-by: Quentin Anthony <[email protected]> Co-authored-by: github-actions <[email protected]> Co-authored-by: lintangsutawika <lintang@stella-ord-0.stella-ord.tenant-eleutherai.svc.tenant.chi.local> Co-authored-by: lintangsutawika <[email protected]> Co-authored-by: dmahan93 <[email protected]> Co-authored-by: aurelion-source <[email protected]> Co-authored-by: aurelion-source <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 836aefa - Browse repository at this point
Copy the full SHA 836aefaView commit details -
Configuration menu - View commit details
-
Copy full SHA for 01e74f4 - Browse repository at this point
Copy the full SHA 01e74f4View commit details -
TE Import Hotfix (EleutherAI#1272)
* fix the te import * refactor get_params_for_weight_decay_optimization * remove incorrect type hint and dead imports
Configuration menu - View commit details
-
Copy full SHA for 61a3daa - Browse repository at this point
Copy the full SHA 61a3daaView commit details -
Add Reward Model training (EleutherAI#1246)
* Add a chat data preprocessing script * add EOT at end of a chat * - add different packing impl (Unpacked, packing until overflow) - fix labels to also have valid/test implementations - fix label masking in _get_batch to also include anything from get_ltor_masks_and_position_ids * update README.md * - Add metrics to forward step to add DPO specific metrics that are useful (accuracy, etc) - Add reference model setup for DPO - Add pairwise dataset for positive/negative pairs - Add DPO loss * Update arguments.py to use train_label_data_paths instead of label_data_paths * - Bugfixes from upstreaming.... * - add precompute logprobs... * - Finishing up precompute logprobs... * - update readme for DPO... * - Add RM training * add comment on why row-parallel for RMs * fix var name --------- Co-authored-by: Quentin Anthony <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 1c72742 - Browse repository at this point
Copy the full SHA 1c72742View commit details -
Configuration menu - View commit details
-
Copy full SHA for bf8e78c - Browse repository at this point
Copy the full SHA bf8e78cView commit details -
Configuration menu - View commit details
-
Copy full SHA for 227967f - Browse repository at this point
Copy the full SHA 227967fView commit details -
Configuration menu - View commit details
-
Copy full SHA for 2a3513c - Browse repository at this point
Copy the full SHA 2a3513cView commit details -
Add configuration options for customizing the experiment name, tags a…
…nd others fields
Configuration menu - View commit details
-
Copy full SHA for 7609829 - Browse repository at this point
Copy the full SHA 7609829View commit details -
Configuration menu - View commit details
-
Copy full SHA for 39f9142 - Browse repository at this point
Copy the full SHA 39f9142View commit details -
Configuration menu - View commit details
-
Copy full SHA for 43ed6e8 - Browse repository at this point
Copy the full SHA 43ed6e8View commit details -
Configuration menu - View commit details
-
Copy full SHA for 90c499a - Browse repository at this point
Copy the full SHA 90c499aView commit details -
Configuration menu - View commit details
-
Copy full SHA for 913f877 - Browse repository at this point
Copy the full SHA 913f877View commit details -
Configuration menu - View commit details
-
Copy full SHA for a6bddd6 - Browse repository at this point
Copy the full SHA a6bddd6View commit details -
Configuration menu - View commit details
-
Copy full SHA for 0468dae - Browse repository at this point
Copy the full SHA 0468daeView commit details -
Configuration menu - View commit details
-
Copy full SHA for ef32d69 - Browse repository at this point
Copy the full SHA ef32d69View commit details -
Configuration menu - View commit details
-
Copy full SHA for 976cd5d - Browse repository at this point
Copy the full SHA 976cd5dView commit details -
Configuration menu - View commit details
-
Copy full SHA for 962314e - Browse repository at this point
Copy the full SHA 962314eView commit details -
Configuration menu - View commit details
-
Copy full SHA for f0a4b70 - Browse repository at this point
Copy the full SHA f0a4b70View commit details -
Configuration menu - View commit details
-
Copy full SHA for c6681b5 - Browse repository at this point
Copy the full SHA c6681b5View commit details -
Configuration menu - View commit details
-
Copy full SHA for 4f76e0d - Browse repository at this point
Copy the full SHA 4f76e0dView commit details