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": [
-    "up_proj",
-    "gate_proj",
     "v_proj",
-    "k_proj",
-    "q_proj",
+    "gate_proj",
+    "up_proj",
     "o_proj",
-    "down_proj"
+    "down_proj",
+    "k_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:f2c0fc8ff8a1ebef0600a00c1892827abe300ebedb0f41ee38becf7a36d29b3a
+oid sha256:b76f4d08d37c0243c957f856f6bbf97f472f6809cabaefaec9bf79fad497b6cc
 size 119975992

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:110acf8b0145f3c61a263e4983634896bf5982c5b50a03f171a93d76b53a6daa
+oid sha256:89886ce704072de2bc7c2e3ad054daeb79eb2d96d797bea29cd5fcc4323e6ec1
 size 5560