Bug fixes#11
Merged
Merged
Conversation
amrothemich
added a commit
to amrothemich/unsloth-zoo
that referenced
this pull request
Nov 6, 2025
KEY DISCOVERY: cache_position corruption starts early with 1156-element tensors instead of single positions. This spreads and corrupts GPU memory. - Detect corrupted cache_position shapes early (call unslothai#11) - Fix by extracting last position and creating proper tensor - Robust flex_attention fallbacks when GPU is corrupted - CPU tensor fallbacks when even torch.ones() fails on GPU This should prevent the CUDA illegal memory access by fixing the root cause before it spreads. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>
mmathew23
added a commit
to mmathew23/unsloth-zoo
that referenced
this pull request
May 22, 2026
Multi-reviewer pass on the autocast wrapper / norm-upcast path: - Instance-level forward (#2): an instance attribute `model.forward` (Unsloth runtime forward patching) shadows class-method overrides, so mutating __class__ silently bypassed the wrapper -> fp32 norm met a bf16 linear with no autocast and crashed. Now wrap the instance attribute when present; otherwise subclass as before. - Wrapper gating (unslothai#5, unslothai#7): install the wrapper iff fp32 norm params actually exist (from our upcast, the legacy env upcast, or an external _pre_set_compute_dtype policy) -- not on the upcast DECISION. Fixes the rollback path leaving external fp32 norms exposed, and stops wrapping models with no fp32 norm. Add _unwrap_forward_in_bf16_autocast for re-prepare (unslothai#10). - config.architectures leak (unslothai#8/unslothai#9): keep the original __name__ on the generated subclass (unique __qualname__ for registration) so save_pretrained records the base architecture. - Device detection (unslothai#11): recurse into mapping/list/tuple batches and fall back to the model's parameter device instead of defaulting to "cuda". - Legacy UNSLOTH_UPCAST_LAYERNORM (#1/#3/unslothai#4): route through the shared _cast_named_module + union matcher and honour the external-policy deferral. - Recursive external-ownership guard (unslothai#6): record descendants of tagged modules (the external policy casts recursively). - Fresh-interpreter pickle test (unslothai#12): real subprocess load. Shared helpers: _find_tensor_device_type, _call_forward_with_bf16_autocast, _canonical_module_name, _cast_named_module. Unit suite: 25 passed.
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.
No description provided.