Training in progress, step 500

adapter_config.json

@@ -1,6 +1,9 @@
 {
   "alpha_pattern": {},
-  "auto_mapping": null,
+  "auto_mapping": {
+    "base_model_class": "LlamaForCausalLM",
+    "parent_library": "transformers.models.llama.modeling_llama"
+  },
   "base_model_name_or_path": "deepseek-ai/deepseek-coder-1.3b-base",
   "bias": "none",
   "fan_in_fan_out": false,
@@ -20,15 +23,15 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
+    "gate_proj",
+    "o_proj",
+    "down_proj",
     "k_proj",
     "v_proj",
     "up_proj",
-    "o_proj",
-    "down_proj",
-    "gate_proj"
+    "q_proj"
   ],
-  "task_type": "CAUSAL_LM",
+  "task_type": null,
   "use_dora": false,
   "use_rslora": false
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7de773cbcc7142be2fb2c20adf30627790ebbff26a4b2ab06acc92d97cd9af8d
+oid sha256:3bf6fb5e046ebde2e80b4f273cff6e55f09a3af387226186773144640e0f38b1
 size 119975992

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47aa4b5956325a412648c53829d88d31b8848bd850a1057c7e4c78d6c3edd6e0
+oid sha256:a6224754869d6640da1463705d3c4acd50b6b9e39f1c75d416bd3be4cb198cad
 size 5560