Training in progress, step 60

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 4.0,
     "eval_accuracy": 1.0,
-    "eval_loss": 0.02880885638296604,
-    "eval_runtime": 1.9696,
-    "eval_samples_per_second": 29.448,
-    "eval_steps_per_second": 4.062,
+    "eval_loss": 0.02930055744946003,
+    "eval_runtime": 2.166,
+    "eval_samples_per_second": 26.777,
+    "eval_steps_per_second": 3.693,
     "total_flos": 1.7978141590585344e+16,
-    "train_loss": 0.325518798828125,
-    "train_runtime": 22.8361,
-    "train_samples_per_second": 10.159,
-    "train_steps_per_second": 2.627
+    "train_loss": 0.3378882090250651,
+    "train_runtime": 22.8868,
+    "train_samples_per_second": 10.137,
+    "train_steps_per_second": 2.622
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
     "eval_accuracy": 1.0,
-    "eval_loss": 0.02880885638296604,
-    "eval_runtime": 1.9696,
-    "eval_samples_per_second": 29.448,
-    "eval_steps_per_second": 4.062
+    "eval_loss": 0.02930055744946003,
+    "eval_runtime": 2.166,
+    "eval_samples_per_second": 26.777,
+    "eval_steps_per_second": 3.693
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3b82fd11abf8d02188081717c815298ef3e6058b0b4c93d3eb091af5dfbb66cc
+oid sha256:953fd1dc14bcfe4abbbc4074f5f494a5910fdeb3f3c545442210eccd9e7d17fe
 size 343223968

runs/Aug30_08-26-44_7e2de4304170/events.out.tfevents.1725006458.7e2de4304170

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad9c4b85611803c1f263162539c266d4c8b44a573e66f834ea13f47c6679544b
+size 357

runs/Aug30_08-27-51_7e2de4304170/events.out.tfevents.1725006472.7e2de4304170

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32f34170112b4fd8d7ec18ca3c8b3d117bdaf93031a56379a2b547dfd4820e12
+size 6439

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
     "total_flos": 1.7978141590585344e+16,
-    "train_loss": 0.325518798828125,
-    "train_runtime": 22.8361,
-    "train_samples_per_second": 10.159,
-    "train_steps_per_second": 2.627
+    "train_loss": 0.3378882090250651,
+    "train_runtime": 22.8868,
+    "train_samples_per_second": 10.137,
+    "train_steps_per_second": 2.622
 }

trainer_state.json

@@ -10,54 +10,54 @@
   "log_history": [
     {
       "epoch": 0.6666666666666666,
-      "grad_norm": 1.813136100769043,
+      "grad_norm": 1.332340121269226,
       "learning_rate": 0.0001666666666666667,
-      "loss": 0.6367,
+      "loss": 0.6343,
       "step": 10
     },
     {
       "epoch": 1.3333333333333333,
-      "grad_norm": 3.1886422634124756,
+      "grad_norm": 2.4813578128814697,
       "learning_rate": 0.00013333333333333334,
-      "loss": 0.7056,
+      "loss": 0.6705,
       "step": 20
     },
     {
       "epoch": 2.0,
-      "grad_norm": 10.936144828796387,
+      "grad_norm": 12.414591789245605,
       "learning_rate": 0.0001,
-      "loss": 0.3764,
+      "loss": 0.4595,
       "step": 30
     },
     {
       "epoch": 2.6666666666666665,
-      "grad_norm": 0.5728101134300232,
+      "grad_norm": 0.5009147524833679,
       "learning_rate": 6.666666666666667e-05,
-      "loss": 0.1584,
+      "loss": 0.1393,
       "step": 40
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 0.2492317408323288,
+      "grad_norm": 0.22140513360500336,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.0441,
+      "loss": 0.0828,
       "step": 50
     },
     {
       "epoch": 4.0,
-      "grad_norm": 0.311509370803833,
+      "grad_norm": 0.21809758245944977,
       "learning_rate": 0.0,
-      "loss": 0.0321,
+      "loss": 0.0409,
       "step": 60
     },
     {
       "epoch": 4.0,
       "step": 60,
       "total_flos": 1.7978141590585344e+16,
-      "train_loss": 0.325518798828125,
-      "train_runtime": 22.8361,
-      "train_samples_per_second": 10.159,
-      "train_steps_per_second": 2.627
+      "train_loss": 0.3378882090250651,
+      "train_runtime": 22.8868,
+      "train_samples_per_second": 10.137,
+      "train_steps_per_second": 2.622
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f0eccb395e989c37741904dbf85ac4c1f348abb3cc07782f6f2135615dffa661
+oid sha256:ba7eb858551fcb097ddfd61ed2f90174151a8d0add0067132e152ce53542b915
 size 5176