Training in progress, step 60

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 4.0,
     "eval_accuracy": 1.0,
-    "eval_loss": 0.02930055744946003,
-    "eval_runtime": 2.166,
-    "eval_samples_per_second": 26.777,
-    "eval_steps_per_second": 3.693,
+    "eval_loss": 0.029831457883119583,
+    "eval_runtime": 1.8404,
+    "eval_samples_per_second": 31.515,
+    "eval_steps_per_second": 4.347,
     "total_flos": 1.7978141590585344e+16,
-    "train_loss": 0.3378882090250651,
-    "train_runtime": 22.8868,
-    "train_samples_per_second": 10.137,
-    "train_steps_per_second": 2.622
+    "train_loss": 0.3446396509806315,
+    "train_runtime": 22.8779,
+    "train_samples_per_second": 10.141,
+    "train_steps_per_second": 2.623
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
     "eval_accuracy": 1.0,
-    "eval_loss": 0.02930055744946003,
-    "eval_runtime": 2.166,
-    "eval_samples_per_second": 26.777,
-    "eval_steps_per_second": 3.693
+    "eval_loss": 0.029831457883119583,
+    "eval_runtime": 1.8404,
+    "eval_samples_per_second": 31.515,
+    "eval_steps_per_second": 4.347
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:953fd1dc14bcfe4abbbc4074f5f494a5910fdeb3f3c545442210eccd9e7d17fe
+oid sha256:e76f6a0c3f5d8ebc533070a28a26b4a506940275f2cb3f217bd38ffa3a1d0c00
 size 343223968

runs/Aug30_08-27-51_7e2de4304170/events.out.tfevents.1725006516.7e2de4304170

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5dde671809bfb4942efed02c7f3bc2798413ead150db9a5be474ceefebf1932
+size 357

runs/Aug30_08-34-08_7e2de4304170/events.out.tfevents.1725006849.7e2de4304170

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40f281a23128f51cc7b3d653bc5cad34c725c99e2757fa2ab8e969b3186db108
+size 6439

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
     "total_flos": 1.7978141590585344e+16,
-    "train_loss": 0.3378882090250651,
-    "train_runtime": 22.8868,
-    "train_samples_per_second": 10.137,
-    "train_steps_per_second": 2.622
+    "train_loss": 0.3446396509806315,
+    "train_runtime": 22.8779,
+    "train_samples_per_second": 10.141,
+    "train_steps_per_second": 2.623
 }

trainer_state.json

@@ -10,54 +10,54 @@
   "log_history": [
     {
       "epoch": 0.6666666666666666,
-      "grad_norm": 1.332340121269226,
+      "grad_norm": 0.9146473407745361,
       "learning_rate": 0.0001666666666666667,
-      "loss": 0.6343,
+      "loss": 0.636,
       "step": 10
     },
     {
       "epoch": 1.3333333333333333,
-      "grad_norm": 2.4813578128814697,
+      "grad_norm": 2.536752700805664,
       "learning_rate": 0.00013333333333333334,
-      "loss": 0.6705,
+      "loss": 0.6707,
       "step": 20
     },
     {
       "epoch": 2.0,
-      "grad_norm": 12.414591789245605,
+      "grad_norm": 8.357431411743164,
       "learning_rate": 0.0001,
-      "loss": 0.4595,
+      "loss": 0.448,
       "step": 30
     },
     {
       "epoch": 2.6666666666666665,
-      "grad_norm": 0.5009147524833679,
+      "grad_norm": 0.41346922516822815,
       "learning_rate": 6.666666666666667e-05,
-      "loss": 0.1393,
+      "loss": 0.1903,
       "step": 40
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 0.22140513360500336,
+      "grad_norm": 0.2532810568809509,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.0828,
+      "loss": 0.0844,
       "step": 50
     },
     {
       "epoch": 4.0,
-      "grad_norm": 0.21809758245944977,
+      "grad_norm": 0.21754373610019684,
       "learning_rate": 0.0,
-      "loss": 0.0409,
+      "loss": 0.0384,
       "step": 60
     },
     {
       "epoch": 4.0,
       "step": 60,
       "total_flos": 1.7978141590585344e+16,
-      "train_loss": 0.3378882090250651,
-      "train_runtime": 22.8868,
-      "train_samples_per_second": 10.137,
-      "train_steps_per_second": 2.622
+      "train_loss": 0.3446396509806315,
+      "train_runtime": 22.8779,
+      "train_samples_per_second": 10.141,
+      "train_steps_per_second": 2.623
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba7eb858551fcb097ddfd61ed2f90174151a8d0add0067132e152ce53542b915
+oid sha256:9afae5936ded42ee4b1a55ec163bc421cd49c1502968a260b3c94b7dde3d5ec0
 size 5176