Finetuning (Trained with Unsloth)

Upload model trained with Unsloth 2x faster

adapter_config.json

@@ -1,7 +1,7 @@
 {
   "alpha_pattern": {},
   "auto_mapping": null,
-  "base_model_name_or_path": "/home/ptelang/test_run/models/Qwen--Qwen2.5-Coder-0.5B",
+  "base_model_name_or_path": "/home/ptelang/curriculum_run/models/Qwen--Qwen2.5-Coder-0.5B",
   "bias": "none",
   "eva_config": null,
   "exclude_modules": null,
@@ -12,24 +12,24 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 64,
+  "lora_alpha": 32,
   "lora_bias": false,
-  "lora_dropout": 0.1,
+  "lora_dropout": 0.0,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 32,
+  "r": 8,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "up_proj",
-    "down_proj",
-    "q_proj",
     "v_proj",
-    "k_proj",
-    "gate_proj"
+    "q_proj",
+    "down_proj",
+    "up_proj",
+    "o_proj",
+    "gate_proj",
+    "k_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d74d932e5bba2f27d3c877ef73de30d8d58d311229aa2492a84c534af6ef3d03
-size 70430032
+oid sha256:e5dbf1cad05250a73211b448a28b0add6874e23e42bd84debae67a404cfcdee4
+size 17640136