Fix gptoss_from_pretrained to correctly load HuggingFace weights#4
Open
Fix gptoss_from_pretrained to correctly load HuggingFace weights#4
Conversation
- Update gptoss_normalize_config to map HF config keys (num_local_experts, num_experts_per_tok, nested rope_scaling) to internal names - Rewrite gptoss_hf_weights_remap to: - Use underscore suffix (_blocks/_scales) for MXFP4 weight detection - Remap HF parameter names to model parameter names - Concatenate separate q/k/v projections into combined qkv tensors Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
gptoss_normalize_configto map HuggingFace config keys (num_local_experts→num_experts,num_experts_per_tok→experts_per_token, nestedrope_scaling) to internal namesgptoss_hf_weights_remapto use underscore suffix (_blocks/_scales) for MXFP4 weight detection, remap HF parameter names to model parameter names, and concatenate separate q/k/v projections into combined qkv tensors🤖 Generated with Claude Code