update model

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "train_loss": 0.27864328904585406,
-    "train_runtime": 413.6435,
+    "train_loss": 0.2811542418508819,
+    "train_runtime": 449.7464,
     "train_samples": 17598,
-    "train_samples_per_second": 127.632,
-    "train_steps_per_second": 7.978
+    "train_samples_per_second": 117.386,
+    "train_steps_per_second": 7.337
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b94a2b7dc1b945a12c97f126d710acab219fcdb45dd98e053089eeb5cb23ba9f
+oid sha256:548c1d2c4a14d27c96fe8163e38824d2b8d6f48e3a7f00fdae61ccd0e2036514
 size 709090132

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "train_loss": 0.27864328904585406,
-    "train_runtime": 413.6435,
+    "train_loss": 0.2811542418508819,
+    "train_runtime": 449.7464,
     "train_samples": 17598,
-    "train_samples_per_second": 127.632,
-    "train_steps_per_second": 7.978
+    "train_samples_per_second": 117.386,
+    "train_steps_per_second": 7.337
 }

trainer_state.json

@@ -10,54 +10,54 @@
   "log_history": [
     {
       "epoch": 0.45,
-      "grad_norm": 3.782728910446167,
+      "grad_norm": 4.830384254455566,
       "learning_rate": 4.242424242424243e-05,
-      "loss": 0.532,
+      "loss": 0.5576,
       "step": 500
     },
     {
       "epoch": 0.91,
-      "grad_norm": 3.925285816192627,
+      "grad_norm": 2.6578476428985596,
       "learning_rate": 3.484848484848485e-05,
-      "loss": 0.4013,
+      "loss": 0.4035,
       "step": 1000
     },
     {
       "epoch": 1.36,
-      "grad_norm": 7.782624244689941,
+      "grad_norm": 6.386310577392578,
       "learning_rate": 2.7272727272727273e-05,
-      "loss": 0.281,
+      "loss": 0.2828,
       "step": 1500
     },
     {
       "epoch": 1.82,
-      "grad_norm": 3.6547534465789795,
+      "grad_norm": 1.7881163358688354,
       "learning_rate": 1.9696969696969697e-05,
-      "loss": 0.2454,
+      "loss": 0.2468,
       "step": 2000
     },
     {
       "epoch": 2.27,
-      "grad_norm": 0.9583206176757812,
+      "grad_norm": 2.0737249851226807,
       "learning_rate": 1.2121212121212122e-05,
-      "loss": 0.1696,
+      "loss": 0.1608,
       "step": 2500
     },
     {
       "epoch": 2.73,
-      "grad_norm": 11.225361824035645,
+      "grad_norm": 0.20467181503772736,
       "learning_rate": 4.5454545454545455e-06,
-      "loss": 0.1393,
+      "loss": 0.1321,
       "step": 3000
     },
     {
       "epoch": 3.0,
       "step": 3300,
       "total_flos": 2871828253461180.0,
-      "train_loss": 0.27864328904585406,
-      "train_runtime": 413.6435,
-      "train_samples_per_second": 127.632,
-      "train_steps_per_second": 7.978
+      "train_loss": 0.2811542418508819,
+      "train_runtime": 449.7464,
+      "train_samples_per_second": 117.386,
+      "train_steps_per_second": 7.337
     }
   ],
   "logging_steps": 500,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:33f26567c9db33a95d0e6f1ccfebacde8b5cba77cf3ea0f0c2923835aa7ee837
+oid sha256:3ad5b11cedb8c71b7172a119a539a49b00c9933a10cf45529cf47a37ef6a7c46
 size 4984