Support Hqq and Eetq quantization #900
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
added 2 commits
May 10, 2024 20:20
added 3 commits
May 11, 2024 10:16
hjh0119
changed the title
tastelikefeet
approved these changes
May 11, 2024
tastelikefeet
added a commit
to tastelikefeet/swift
that referenced
this pull request
May 14, 2024
* commit 'bdc8f54848daad335e513183482e16cc5da17c88': (36 commits) fix export self-cognition (modelscope#929) fix deepseek2(modelscope#924) Add 34b quantized model (modelscope#920) yi1.5 quantized model (modelscope#917) update readme&doc (modelscope#916) init (modelscope#915) fix unsloth import (modelscope#912) add more models local repo support (modelscope#911) lint DeepseekVL add local_repo_path argument AND infer support delete truncation_strategy (modelscope#883) Support Hqq and Eetq quantization (modelscope#900) fix val_sample (modelscope#909) Add val_dataset argument (modelscope#906) Refactor sequence parallel (modelscope#823) replace dataset name with modelscope dataset id (modelscope#899) replace dataset name with dataset path from modelscope (modelscope#897) fix doc link enable longlora and adalora merge (modelscope#892) fix lisa show bug (modelscope#891) update doc (modelscope#888) ... # Conflicts: # swift/llm/utils/dataset.py
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
PR type
PR information
New quantization algorithms:
To resolve the garbled output issue during inference with internvl-chat-v1.5-int8 on V100 machines, fix the dtype to bf16.
Add
device_map_config_pathin theSftArgumentas the file path for manually configuring the model device map.Modified the training script for
deepseek-v2-chat