Phi3_8/5

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-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7888
+- Loss: 0.7798
 
 ## Model description
 
@@ -52,10 +52,10 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| No log        | 0.9948 | 24   | 0.8265          |
-| 1.2331        | 1.9896 | 48   | 0.7947          |
-| 0.6288        | 2.9845 | 72   | 0.7857          |
-| 0.5615        | 3.9793 | 96   | 0.7888          |
+| No log        | 0.9948 | 24   | 0.8300          |
+| 1.2284        | 1.9896 | 48   | 0.7882          |
+| 0.6264        | 2.9845 | 72   | 0.7763          |
+| 0.5586        | 3.9793 | 96   | 0.7798          |
 
 
 ### Framework versions

adapter_config.json

@@ -20,10 +20,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "qkv_proj",
     "down_proj",
-    "gate_up_proj"
+    "gate_up_proj",
+    "qkv_proj",
+    "o_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:331bb481e641c50e188cfff4315ba294cd0f3b659be09dbe7a62d40a4c8a0b2c
+oid sha256:4993e640f4381a1aa3993ae3538991330cdb2389e7c8b25e6569298dc9e5052c
 size 100697728

runs/May08_08-54-21_a53ccbe00de5/events.out.tfevents.1715158693.a53ccbe00de5.416.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:317e021f1811c8a88f372230a69f82048b0482224eb517b17bb861b367be1bc3
+size 7515

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:239a529daba4d3e658280bc8234215f97e3ae6d615036f77b170ff483149b0cf
+oid sha256:7388d6afddb9064fc480ce25d5ccf7a87618cbcb0c1c65bf8e0bee669119dd3e
 size 4984