Upload README.md with huggingface_hub

README.md

@@ -7,17 +7,17 @@ tags:
 
 # Mini-Merge-35B-A3B
 
-**This is a model *merge*, not an original training run.** It was produced by a plain
-element-wise weighted average of two existing open-weight checkpoints -- a ~40-line
-safetensors script, no mergekit (its catalog doesn't know this architecture). Zero
-pretraining, zero post-training of our own. The exact recipe ships as `merge.yaml`
-alongside the weights, so the "how" is not a mystery.
+This is a model **merge**, not an original training run. It's a plain element-wise
+weighted average of two existing open-weight checkpoints (a ~40-line safetensors script;
+no mergekit, since its catalog doesn't know this architecture). No pretraining, no
+post-training of our own. The exact recipe ships as `merge.yaml` next to the weights.
 
 ## Ingredients
 - `nex-agi/Nex-N2-mini`
 - `Qwen/Qwen3.6-35B-A3B`
 
-Merge variant: `mini-b0.2`. Full pipeline: the merge-theater repo.
+Merge variant: `mini-b0.2`. Full pipeline and commands:
+https://github.com/PixelML/merge-theater
 
 ## Benchmarks (same lm-eval tasks, parents vs merge)
 
@@ -36,14 +36,18 @@ Aggregate (mean over shared tasks):
 - Qwen3.6-35B-A3B: 61.37
 
 A merge topping a parent by a point or two mostly reflects eval-format luck, not new
-capability. Read accordingly; the numbers are reproducible with the command below.
+capability, so read it that way. The numbers reproduce with the command below.
 
 ## Reproduce
 ```bash
-modal run modal_app.py::sweep      # or ::rematch for the 397B build
+modal run modal_app.py::sweep --weights-b 0.2,0.3,0.4
 ```
 
 ## License
-Both parents (`nex-agi/Nex-N2-mini` and `Qwen/Qwen3.6-35B-A3B`) are Apache-2.0, so this merge is released under
-**Apache-2.0**. The merge used a small original script, not mergekit, so no LGPL applies.
-This card must travel with the weights.
+Both parents (`nex-agi/Nex-N2-mini` and `Qwen/Qwen3.6-35B-A3B`) are Apache-2.0, so this merge is released under Apache-2.0.
+The merge used a small original script, not mergekit, so no LGPL applies. This card must
+travel with the weights.
+
+## Credit
+Built by Claude Code (Anthropic's agent) running autonomously: the merge, the eval
+pipeline, the ratio sweep, and this card.