Model save

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "train_loss": 0.8506233294804891,
-    "train_runtime": 32.2126,
+    "train_loss": 0.8458248774210612,
+    "train_runtime": 38.2645,
     "train_samples": 42,
-    "train_samples_per_second": 1.304,
-    "train_steps_per_second": 0.093
+    "train_samples_per_second": 1.098,
+    "train_steps_per_second": 0.078
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f4133f7d8471ad9c6f30236ce290a19e7335628900703fb5422126f67149d481
+oid sha256:96757bee906119aac800b9f547e51e7c5ce2ccce2d4097f0dc0f59a533e2b3f6
 size 1324830880

runs/Apr24_10-34-49_MacBook-Pro-3.local/events.out.tfevents.1713947696.MacBook-Pro-3.local.89842.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8a8fc63a7c5025b5628c3dbf083ebd194d3378c485af1945885b4f2c7d0e6f9
+size 5828

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "train_loss": 0.8506233294804891,
-    "train_runtime": 32.2126,
+    "train_loss": 0.8458248774210612,
+    "train_runtime": 38.2645,
     "train_samples": 42,
-    "train_samples_per_second": 1.304,
-    "train_steps_per_second": 0.093
+    "train_samples_per_second": 1.098,
+    "train_steps_per_second": 0.078
 }

trainer_state.json

@@ -10,27 +10,27 @@
   "log_history": [
     {
       "epoch": 0.33,
-      "grad_norm": 3.046247720718384,
+      "grad_norm": 10.981945037841797,
       "learning_rate": 5e-07,
-      "logits/chosen": -1.6115068197250366,
-      "logits/rejected": -1.663214921951294,
-      "logps/chosen": -152.84686279296875,
-      "logps/rejected": -212.77816772460938,
-      "loss": 0.6859,
-      "rewards/accuracies": 0.8125,
-      "rewards/chosen": -0.0250315573066473,
-      "rewards/margins": 0.014816472306847572,
-      "rewards/rejected": -0.03984802961349487,
+      "logits/chosen": -0.6707152128219604,
+      "logits/rejected": -0.8389236927032471,
+      "logps/chosen": -465.56787109375,
+      "logps/rejected": -1116.6866455078125,
+      "loss": 1.1531,
+      "rewards/accuracies": 0.4375,
+      "rewards/chosen": 0.2850288152694702,
+      "rewards/margins": 0.3632635474205017,
+      "rewards/rejected": -0.07823469489812851,
       "step": 1
     },
     {
       "epoch": 1.0,
       "step": 3,
       "total_flos": 0.0,
-      "train_loss": 0.8506233294804891,
-      "train_runtime": 32.2126,
-      "train_samples_per_second": 1.304,
-      "train_steps_per_second": 0.093
+      "train_loss": 0.8458248774210612,
+      "train_runtime": 38.2645,
+      "train_samples_per_second": 1.098,
+      "train_steps_per_second": 0.078
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc01db7ce871611d3b2d6d801e8ed59da76ce78617456a1bb8b1d85857e2794b
+oid sha256:f3f7f97f31ef1eb375dd9b7308555e225d5387fb7fe51d12f7410ebb4a4eb480
 size 5048