Training in progress, epoch 1

adapter_config.json

@@ -2,7 +2,7 @@
   "alpha_pattern": {},
   "auto_mapping": null,
   "base_model_name_or_path": "bigscience/bloom-3b",
-  "bias": "lora_only",
+  "bias": "all",
   "fan_in_fan_out": false,
   "inference_mode": true,
   "init_lora_weights": true,
@@ -13,7 +13,10 @@
   "lora_dropout": 0.1,
   "megatron_config": null,
   "megatron_core": "megatron.core",
-  "modules_to_save": null,
+  "modules_to_save": [
+    "qa_outputs.weight",
+    "qa_outputs.bias"
+  ],
   "peft_type": "LORA",
   "r": 16,
   "rank_pattern": {},

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68ff1f3ad8b4b149edaa6a468a41fc11e9435f1bab5ed59223898daf93b57c52
-size 19680196
+oid sha256:092e010c6c99acbd78abbfd1cde7e4135e8f9bd8ac269ef6af8d4624be1d641f
+size 21403940

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d4c56ac89f13363d5aac868d49f97fee18964516ec702585417c21c2e241dbf
+oid sha256:0ca301bdb8f5275e961724687fe4ca74d3f419657165fe8ed959ec5906bf9695
 size 4728