Upload adapter weights as intrinsic.

stembolts/granite-4.0-h-micro/alora/adapter_config.json

@@ -0,0 +1,48 @@
+{
+  "alora_invocation_tokens": [
+    100264,
+    2071,
+    14112,
+    479,
+    100265
+  ],
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "ibm-granite/granite-4.0-h-micro",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj",
+    "k_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

stembolts/granite-4.0-h-micro/alora/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e191af8459dea2c34b1ed0c066b6687a2bece5445e7ae42217f1d8c1a662fdb2
+size 4721800

stembolts/granite-4.0-h-micro/alora/io.yaml

@@ -0,0 +1,24 @@
+# Model name string, or null to use whatever is provided in the chat completion request.
+model: ~
+response_format: |
+  {
+    "properties": {
+      "defective_part": {
+        "title": "name of defective part, or unknown",
+        "type": "string"
+      },
+      "diag_likelihood": {
+        "title": "likelihood of correct detective part identification",
+        "type": "number"
+      }
+    },
+    "required": [
+      "defective_part", "diag_likelihood"
+    ],
+    "title": "Stembolt Detective Part",
+    "type": "object"
+  }
+transformations: null
+parameters:
+  # Current LoRA can be quite verbose in its explanations.
+  max_completion_tokens: 1024