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-1
   results: []
@@ -20,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # phi-2-gpo-test-iter-1
 
-This model is a fine-tuned version of [DUAL-GPO/phi-2-gpo-test-iter-0](https://huggingface.co/DUAL-GPO/phi-2-gpo-test-iter-0) 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": [
-    "k_proj",
     "q_proj",
     "dense",
-    "v_proj"
+    "v_proj",
+    "k_proj"
   ],
   "task_type": "CAUSAL_LM"
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c99237b61ddbcc12b02ca16d9318ff7a9125e77add338ae7faccac130d33f3ad
+oid sha256:e6c65e876e3560feee1c32a18996ea5d2c1ec0cb84af0807dfae3deb2d67210e
 size 41977616

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.0,
-    "train_loss": 0.0008448870503343642,
-    "train_runtime": 3.6021,
+    "epoch": 1.33,
+    "train_loss": 0.0015329468587879092,
+    "train_runtime": 7.2892,
     "train_samples": 30567,
-    "train_samples_per_second": 2.221,
-    "train_steps_per_second": 0.555
+    "train_samples_per_second": 2.744,
+    "train_steps_per_second": 0.274
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.0,
-    "train_loss": 0.0008448870503343642,
-    "train_runtime": 3.6021,
+    "epoch": 1.33,
+    "train_loss": 0.0015329468587879092,
+    "train_runtime": 7.2892,
     "train_samples": 30567,
-    "train_samples_per_second": 2.221,
-    "train_steps_per_second": 0.555
+    "train_samples_per_second": 2.744,
+    "train_steps_per_second": 0.274
 }

trainer_state.json

@@ -1,7 +1,7 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.0,
+  "epoch": 1.3333333333333333,
   "eval_steps": 100,
   "global_step": 2,
   "is_hyper_param_search": false,
@@ -11,11 +11,11 @@
     {
       "epoch": 1.0,
       "learning_rate": 5e-06,
-      "logits/chosen": 0.09949145466089249,
-      "logits/rejected": -0.044053155928850174,
-      "logps/chosen": -259.9786682128906,
-      "logps/rejected": -270.3120422363281,
-      "loss": 0.0008,
+      "logits/chosen": 0.7674216628074646,
+      "logits/rejected": 0.7469885945320129,
+      "logps/chosen": -160.50914001464844,
+      "logps/rejected": -129.6922149658203,
+      "loss": 0.0023,
       "rewards/accuracies": 0.0,
       "rewards/chosen": 0.0,
       "rewards/margins": 0.0,
@@ -23,13 +23,13 @@
       "step": 1
     },
     {
-      "epoch": 2.0,
+      "epoch": 1.33,
       "step": 2,
       "total_flos": 0.0,
-      "train_loss": 0.0008448870503343642,
-      "train_runtime": 3.6021,
-      "train_samples_per_second": 2.221,
-      "train_steps_per_second": 0.555
+      "train_loss": 0.0015329468587879092,
+      "train_runtime": 7.2892,
+      "train_samples_per_second": 2.744,
+      "train_steps_per_second": 0.274
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e96dcce776491cac70e439cf6c0110112c35c2483e211be90e97da63c4f336ef
+oid sha256:1545e3e119ea55ccb2756568d3ff67054bce5657c30f0f821154534a85010e1c
 size 5816