Model save

README.md

@@ -2,14 +2,10 @@
 license: mit
 library_name: peft
 tags:
-- alignment-handbook
-- generated_from_trainer
 - trl
 - dpo
 - generated_from_trainer
 base_model: microsoft/phi-2
-datasets:
-- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: phi-2-gpo-test-iter-0
   results: []
@@ -20,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # phi-2-gpo-test-iter-0
 
-This model is a fine-tuned version of [lole25/phi-2-sft-ultrachat-lora](https://huggingface.co/lole25/phi-2-sft-ultrachat-lora) on the HuggingFaceH4/ultrafeedback_binarized dataset.
+This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on the None dataset.
 
 ## Model description
 

adapter_config.json

@@ -19,10 +19,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "dense",
     "q_proj",
-    "k_proj"
+    "k_proj",
+    "v_proj",
+    "dense"
   ],
   "task_type": "CAUSAL_LM"
 }

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.0,
-    "train_loss": 2.5006682335515507e-05,
+    "train_loss": 0.00012263256940059364,
     "train_runtime": 3.6037,
     "train_samples": 30567,
-    "train_samples_per_second": 1.11,
+    "train_samples_per_second": 2.22,
     "train_steps_per_second": 0.555
 }

runs/Mar19_16-09-02_Braylon/events.out.tfevents.1710825001.Braylon.14685.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8faa91ffa5a41f9a24375a8552fb98f0d743e864b65540a6365ad1da67b4dbca
+size 5793

runs/Mar19_17-53-52_Braylon/events.out.tfevents.1710831296.Braylon.1078.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddd7589d57da76796f7bcca3c485d8d29f2bd7173a5a29ddb02bf250de98700f
+size 5445

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.0,
-    "train_loss": 2.5006682335515507e-05,
+    "train_loss": 0.00012263256940059364,
     "train_runtime": 3.6037,
     "train_samples": 30567,
-    "train_samples_per_second": 1.11,
+    "train_samples_per_second": 2.22,
     "train_steps_per_second": 0.555
 }

trainer_state.json

@@ -11,11 +11,11 @@
     {
       "epoch": 1.0,
       "learning_rate": 5e-06,
-      "logits/chosen": 0.780505895614624,
-      "logits/rejected": 0.9561213254928589,
-      "logps/chosen": -174.01596069335938,
-      "logps/rejected": -72.11927795410156,
-      "loss": 0.0,
+      "logits/chosen": 0.7793408036231995,
+      "logits/rejected": 0.853427529335022,
+      "logps/chosen": -255.33981323242188,
+      "logps/rejected": -306.3643798828125,
+      "loss": 0.0001,
       "rewards/accuracies": 0.0,
       "rewards/chosen": 0.0,
       "rewards/margins": 0.0,
@@ -26,9 +26,9 @@
       "epoch": 2.0,
       "step": 2,
       "total_flos": 0.0,
-      "train_loss": 2.5006682335515507e-05,
+      "train_loss": 0.00012263256940059364,
       "train_runtime": 3.6037,
-      "train_samples_per_second": 1.11,
+      "train_samples_per_second": 2.22,
       "train_steps_per_second": 0.555
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47c95aef615e88954ec6528c8eb6c47f35003d392815c5fd45a10bd196f06354
-size 5880
+oid sha256:6eba0d55a58d5d9aa3a78f3e6b463d6676cdebf1144903cfc1e8dedb97e8a4c4
+size 5816