Training in progress, step 78

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": [
-    "v_proj",
-    "k_proj",
     "q_proj",
-    "gate_proj",
-    "down_proj",
     "up_proj",
-    "o_proj"
+    "k_proj",
+    "down_proj",
+    "gate_proj",
+    "o_proj",
+    "v_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 sha256:4eecd3082d40f9be7e8ed4c12ffc92e029fc7bbb361653edf0fcbc5faeed5e6e
+oid sha256:a19034b6bf8fd643d95c08b29c7d6684a0057e57e98510e0043337403cafab34
 size 119975992

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:19bb51975b6823bbdff4d7541e79ef087865c63bf95705990a70056564ad7e3b
+oid sha256:808c68d69e320211cf621847e96ab8b624a67d4fe3a45514b384b891a9279bdf
 size 5560