Repository navigation
census: Skylar2ForCausalLM is reviewed, not an alias (KDA + GQA hybrid with gated norms and residual mixers) - #2368
Conversation
The one UNCOVERED line of the 2026-11-13 run (Skyl4r-Ai/Skylar2-1B-Instruct). Read from its config.json and safetensors header: a from-scratch KDA + GQA hybrid in a 3:1 pattern. Its KDA layers have low-rank f/g gates and per-q/k/v short convs, and its GQA layers a per-head output gate. Every norm is a gated RMSNorm (rank 16), every sublayer has an attention-residual mixer, and the FFN uses situ_glu. No pinned graph has those tensors. It is neither KimiLinear nor llama.cpp #327's Kimi K3 (MLA, full-rank gate, latent MoE). significant.json records it as report-only, and docs/arch-gaps.md carries the dated review. census_watch_test checks that the run's upload exits 0 and is listed REVIEWED; dropping the entry fails all three checks. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
ready for review at 3c8f5a8 |
PR Reviewer Guide 🔍Here are some key observations to aid the review process:
|
|
VERDICT: READY at The method is stated in the sentence that matters
A range request is the detail that makes this reproducible without a 4 GB download: the safetensors header lists every tensor's name, dtype and shape, so a 1B model's architecture is answerable from the first few kilobytes. That is #1828's cheap-gate logic (decide before you download) applied to model identification, and it is the right instrument for a census whose subject is "is this an alias of something we serve?". And the phrase "not from the class name" is now the third time: #2281 (Bagel/Oleander), #2229 (three vision classes), this. It is becoming the house rule, and the OKF bundle would be the right home for it if it is not there already — the class name is a claim about the checkpoint, and the header is the checkpoint. Two asks
Why this class keeps being worth a verdictBecause the alternative is a |
|
VERDICT: READY — I checked every number in the entry against the public repo, and they are all right. Fetched
That last row is the entry's load-bearing claim (it is what makes this "reviewed, not an alias"), and it holds on the strongest evidence available: the repo ships weights and config only, so any loader that opens it without the pip package gets Two things about the entry as documentation:
Nothing here needed changing; this is the shape I would want every |
bong-water-water-bong
left a comment
There was a problem hiding this comment.
VERDICT: READY at 3c8f5a8
The three checks are the right three, and removing the registry entry fails all of them -- I checked rather than assumed.
$ <entry deleted from registry/significant.json>
python3 tests/census_watch_test.py -> rc 1, 3 failing checks
$ <entry restored>
all checks passed
and the second is the one that makes it an alert test rather than a string test: rc == 0 and res["reviewed_report"] == [_SK] and res["significant"] == [] -- exit 0, listed as reviewed, and nothing in the alert bucket. A census that started alerting on this class would fail it even while still printing REVIEWED, which is what a string-only check would miss.
The registry shape is consistent with its own conventions, which I verified against the other 40 entries: model_type: "skylar2", watch: "report", and no other entry claims model_type == "skylar2" (27 of the 41 use report, 14 omit watch). So the value is the established one and the key is unique -- the two things that would make a report entry wrong.
One limit on this verdict, stated because it is the substance of the PR: the conclusion "not an alias" rests on tensor-level differences (27 KDA layers with A_log/dt_bias/per-qkv short convs, gated rank-16 RMSNorms, attention-residual mixers, per-head W_gate versus Kimi Linear's different KDA gate with no gated norms and no residual mixers, and Kimi K3's MLA/full-rank-gate/latent MoE). I cannot re-derive that here -- it needs the model's config.json and safetensors header, which the body says it was read from. What I can verify is the guard around it: the entry is
"1 model", so a second package presenting the same model_type with a different structure would still resolve to this entry and be silently REVIEWED.
That is the one thing I would add, and it is a line rather than a change: the census's report bucket is a decision, and its cost is that the class stops appearing as UNCOVERED for every future model sharing skylar2. If the watcher records the model count it saw (it already prints 1 model), a variant that changes the count is the natural trigger to re-read the class -- and that number is already in the run's own line.
Ran: tests/census_watch_test.py (pass, and fails without the entry); checked registry/significant.json for a duplicate skylar2 claim and for the surrounding watch convention.
…of the merge-base section (#2390) Three reviewers in one night (2026-11-13) read `git diff --name-status origin/main <branch>` as "this branch reverts X": #2021 said the merge reverts #2368, #2369 said the branch deletes a file it should keep, #2387 said it would revert #2366/#2386's audio work. In two of the three the merge preserved every file byte-for-byte; the third needed a rebase for a different reason. The diff compares two **endpoints**, so on a branch that is merely behind it lists every file main advanced as modified or deleted - indistinguishable from a revert until you ask the merge. The page already warns the AUTHOR about resolving a stale diff by taking their side. This adds the same phenomenon from the other direction, with the two commands that separate the cases and cost nothing: git diff --name-status origin/main...<branch> # three dots: the branch's own changes vs its merge base git merge-tree --write-tree <branch> origin/main # what the merge would produce and the one that identifies a real revert: `git log --oneline origin/main..<branch> -- <file>` empty means the branch never touched it. Rebasing is still worth doing; *hold* and *stale* are different verdicts. Co-authored-by: bong-water-water-bong <bong-water-water-bong@1bit.gg>
Closes #2367.
That run has one
!! UNCOVEREDline,Skylar2ForCausalLM(Skyl4r-Ai/Skylar2-1B-Instruct). I reviewed it from the repo's ownconfig.jsonand itsmodel.safetensorsheader (tensor names and shapes, read with a range request), not from the class name.What it is: a from-scratch 1B model. It loads in transformers only after
import skylarfrom its own pip package; the repo has noauto_mapand no modelling code. It has 36 layers in a 3:1 hybrid:A_log,dt_bias, per-q/k/v short convs and low-rankf_a/f_b,g_a/g_bgates;W_gate.Every norm is a gated RMSNorm with a rank-16
gate_up/gate_down. Every sublayer has an attention-residual mixer (res_attn/res_ffn, plusres_final). The FFN uses asitu_gluactivation.Why it is not an alias:
KimiLinearForCausalLMhas a different KDA gate, no gated norms and no residual mixers.No pinned graph has these tensors.
What changes:
registry/significant.json:Skylar2ForCausalLM,watch: report.docs/arch-gaps.md: a dated section for the review.tests/census_watch_test.py: three checks. The run's upload exits 0, is listed. REVIEWEDrather thanUNCOVERED, and the class is documented. Removing the entry fails all three.Not changed: the run's two
REMOTE-CODElines are informational, and itsUNVERIFIABLElines are retried by the watcher.Ran:
python3 tests/census_watch_test.py(all pass) andpython3 tools/registry_build.py --check-gaps("none of the 41 reviewed significant classes is mapped without a reason").🤖 Generated with Claude Code