FinaPolat/microsoft-phi-3.5-multidomain-ner-qlora-adapter

README.md

@@ -18,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/Phi-3.5-mini-instruct](https://huggingface.co/microsoft/Phi-3.5-mini-instruct) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3157
+- Loss: 0.3230
 
 ## Model description
 
@@ -46,7 +46,7 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: constant
 - lr_scheduler_warmup_ratio: 0.3
-- num_epochs: 2
+- num_epochs: 1
 
 ### Training results
 

adapter_config.json

@@ -21,10 +21,10 @@
   "revision": null,
   "target_modules": [
     "q_proj",
-    "gate_proj",
-    "v_proj",
+    "o_proj",
     "k_proj",
-    "o_proj"
+    "v_proj",
+    "gate_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0eec16b70583286d0e9d01b45226815dac47edff5c10e5637a3115b4c00c2463
+oid sha256:fe30a3fc07b9322648f594bf27d2669cb968bb84b028ddc2b7f1f921a60bbf88
 size 25174424

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:36cd7bbc9fa7878a474498ce728158cd335fc50e139540341dd078e7c4f44758
-size 5496
+oid sha256:acadda6863e07b08849474c95189b5dc66478f06486f868d140a7dd9cb7332dd
+size 5560