Publish audited portallib v0.1.0 artifact

README.md

@@ -0,0 +1,67 @@
+---
+library_name: portallib
+license: apache-2.0
+base_model: Qwen/Qwen3-8B
+datasets:
+- RampPublic/portallib-tasks
+tags:
+- portal
+- hypernetwork
+- lora
+- peft
+- multiple-choice
+---
+
+# PorTAL refit for Qwen3-8B
+
+This is a native [PorTAL](https://github.com/ramp-public/portallib) artifact refitted onto
+`Qwen/Qwen3-8B`. Its 14-task latent table and canonical LoRA-generating core were learned jointly
+from Qwen3-1.7B and Qwen3-4B and frozen during refitting. Only a fresh Qwen3-8B alignment was
+trained. The artifact generates rank-8 LoRA factors for the query and value projections of every
+decoder layer.
+
+## Evaluation
+
+One seed was evaluated on the complete 14-task validation suite using continuation log-probability
+divided by character length (`acc_norm`). Gold continuation token-mean NLL was tracked separately
+for checkpoint selection.
+
+| Model | Macro `acc_norm` |
+|---|---:|
+| Frozen Qwen3-8B | 0.6681 |
+| PorTAL-adapted | 0.7767 |
+| Absolute lift | +0.1086 |
+
+These are research benchmark results for this exact artifact and evaluation recipe, not a general
+performance guarantee.
+
+## Refit recipe
+
+- Base: `Qwen/Qwen3-8B` at `b968826d9c46dd6066d109eabc6255188de91218`
+- Frozen source carrier: `RampPublic/portal-qwen3-4b`
+- Dataset: `RampPublic/portallib-tasks` at
+  `d35f1e8a813cfae662166164fc25965a31b01ae0`
+- Refit data: deterministic seeded sample of up to 1,000 examples per task from the complete
+  training pool
+- Optimization: 5 epochs, batch size 4, alignment LR `1e-3`, linear decay with 10% warmup, seed 0
+- Trainable parameters: target-base alignment only; task latents and canonical core remain frozen
+- Checkpoint: maximum macro validation `acc_norm`, with lower gold NLL as the tie-breaker
+- Architecture: q/v targets, rank 8, alpha 16, task latent 256, layer embedding 32, hidden 512,
+  canonical width 1024
+
+## Usage
+
+```python
+from portallib import PortalModel
+
+portal = PortalModel.from_pretrained(
+    "RampPublic/portal-qwen3-8b",
+    revision="v0.1.0",
+)
+portal.export_peft("rte", "./portal-rte-qwen3-8b")
+```
+
+See the [release recipe](https://github.com/ramp-public/portallib/blob/main/REPRODUCING.md) for the
+full task list, evaluation definition, and refitting procedure. The artifact is Apache-2.0; the
+benchmark dataset contains components under multiple upstream licenses documented on its dataset
+card.

config.json

@@ -0,0 +1,48 @@
+{
+  "alpha": 16,
+  "architecture": "canonical",
+  "base_model_name_or_path": "Qwen/Qwen3-8B",
+  "base_model_revision": "b968826d9c46dd6066d109eabc6255188de91218",
+  "d_core": 1024,
+  "d_layer": 32,
+  "d_z": 256,
+  "hidden": 512,
+  "in_dims": {
+    "q": 4096,
+    "v": 4096
+  },
+  "layer_path": "model.layers",
+  "library_name": "portallib",
+  "module_paths": {
+    "q": "self_attn.q_proj",
+    "v": "self_attn.v_proj"
+  },
+  "n_layers": 36,
+  "out_dims": {
+    "q": 4096,
+    "v": 1024
+  },
+  "rank": 8,
+  "schema_version": 1,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "tasks": [
+    "truthfulqa",
+    "rte",
+    "cb",
+    "copa",
+    "wic",
+    "wsc",
+    "boolq",
+    "arc_easy",
+    "arc_challenge",
+    "hellaswag",
+    "openbookqa",
+    "winogrande",
+    "commonsense_qa",
+    "sciq"
+  ]
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f9d42d6431fe0a8f201df8d36f841cf1151690690794ff16fb273fd193b7de0
+size 124481816