server-train: no held-out tokens, no eval loss (null, not 0.0); report eval_trainable_tokens - #31
Merged
joelteply merged 1 commit intoSep 28, 2026
Conversation
… eval_trainable_tokens reports the count With val_split 0, or a split that rounds to no windows, trainable_eval is 0, scale_eval is 0, and ggml-opt's empty eval result gives loss 0.0. The epoch then reported eval_loss 0.0, and the core read that as a held-out validation loss of zero (Codex). Now eval_loss is null when no held-out token was evaluated, and state carries eval_trainable_tokens, so a consumer knows whether a held-out measurement exists. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LoTjvf5j3Ez13g6k8mRkFo
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
With val_split 0, or a split that rounds to no held-out windows, trainable_eval is 0. ggml-opt's empty eval result then gives loss 0.0, and the epoch reported eval_loss 0.0. The core read that as a held-out validation loss of zero (Codex, from source: server-train.cpp 689-709, ggml-opt.cpp 661).
The core side reads both (continuum PR to follow).
🤖 Generated with Claude Code
https://claude.ai/code/session_01LoTjvf5j3Ez13g6k8mRkFo