Model save

README.md

@@ -27,7 +27,7 @@ print(output["generated_text"])
 
 ## Training procedure
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/yh-yao/huggingface/runs/5kvjtgpr) 
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/yh-yao/huggingface/runs/8bndsvn7) 
 
 
 This model was trained with SFT.

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
     "total_flos": 76916824473600.0,
-    "train_loss": 0.8026524422436776,
-    "train_runtime": 501.113,
+    "train_loss": 0.8026537260360266,
+    "train_runtime": 501.1144,
     "train_samples": 16610,
     "train_samples_per_second": 43.124,
     "train_steps_per_second": 0.337

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af00eb94ed124448f9d00eef728b710075389ebd059cafb1d96fc028a1ddf502
+oid sha256:76efe68980ef11b7ac637b0710c0cbf60cb19d0903a4c7dd0fc77e6f30b24050
 size 3087467144

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
     "total_flos": 76916824473600.0,
-    "train_loss": 0.8026524422436776,
-    "train_runtime": 501.113,
+    "train_loss": 0.8026537260360266,
+    "train_runtime": 501.1144,
     "train_samples": 16610,
     "train_samples_per_second": 43.124,
     "train_steps_per_second": 0.337

trainer_state.json

@@ -10,239 +10,239 @@
   "log_history": [
     {
       "epoch": 0.029585798816568046,
-      "grad_norm": 2.3454635635521646,
+      "grad_norm": 2.3455099607931436,
       "learning_rate": 5.882352941176471e-06,
       "loss": 1.0993,
       "step": 5
     },
     {
       "epoch": 0.05917159763313609,
-      "grad_norm": 1.636082977147667,
+      "grad_norm": 1.6331238734695321,
       "learning_rate": 1.1764705882352942e-05,
-      "loss": 1.0403,
+      "loss": 1.0402,
       "step": 10
     },
     {
       "epoch": 0.08875739644970414,
-      "grad_norm": 0.8858412348123735,
+      "grad_norm": 0.8886649764083925,
       "learning_rate": 1.7647058823529414e-05,
       "loss": 0.9533,
       "step": 15
     },
     {
       "epoch": 0.11834319526627218,
-      "grad_norm": 0.6060211903989927,
+      "grad_norm": 0.6057426213937779,
       "learning_rate": 1.9980782984658682e-05,
       "loss": 0.8923,
       "step": 20
     },
     {
       "epoch": 0.14792899408284024,
-      "grad_norm": 0.5401629650362618,
+      "grad_norm": 0.5392641409351531,
       "learning_rate": 1.9863613034027224e-05,
       "loss": 0.8552,
       "step": 25
     },
     {
       "epoch": 0.17751479289940827,
-      "grad_norm": 0.42641787400688946,
+      "grad_norm": 0.4265970897513392,
       "learning_rate": 1.9641197940012136e-05,
       "loss": 0.8283,
       "step": 30
     },
     {
       "epoch": 0.20710059171597633,
-      "grad_norm": 0.3766738990387521,
+      "grad_norm": 0.3763318790920631,
       "learning_rate": 1.9315910880512792e-05,
-      "loss": 0.8229,
+      "loss": 0.823,
       "step": 35
     },
     {
       "epoch": 0.23668639053254437,
-      "grad_norm": 0.3996293636278583,
+      "grad_norm": 0.4001698330922381,
       "learning_rate": 1.8891222681391853e-05,
       "loss": 0.8225,
       "step": 40
     },
     {
       "epoch": 0.26627218934911245,
-      "grad_norm": 0.34546697048235825,
+      "grad_norm": 0.34301763196632185,
       "learning_rate": 1.8371664782625287e-05,
       "loss": 0.8073,
       "step": 45
     },
     {
       "epoch": 0.2958579881656805,
-      "grad_norm": 0.3349879831536363,
+      "grad_norm": 0.33452509246677203,
       "learning_rate": 1.7762780887657576e-05,
       "loss": 0.7977,
       "step": 50
     },
     {
       "epoch": 0.3254437869822485,
-      "grad_norm": 0.36751019334772667,
+      "grad_norm": 0.3655532785130365,
       "learning_rate": 1.7071067811865477e-05,
       "loss": 0.7877,
       "step": 55
     },
     {
       "epoch": 0.35502958579881655,
-      "grad_norm": 0.36912550445461845,
+      "grad_norm": 0.371555208555247,
       "learning_rate": 1.6303906161279554e-05,
       "loss": 0.7981,
       "step": 60
     },
     {
       "epoch": 0.38461538461538464,
-      "grad_norm": 0.3411495577853807,
+      "grad_norm": 0.34243630022925536,
       "learning_rate": 1.5469481581224274e-05,
       "loss": 0.7722,
       "step": 65
     },
     {
       "epoch": 0.41420118343195267,
-      "grad_norm": 0.34837356713937606,
+      "grad_norm": 0.34747447939093157,
       "learning_rate": 1.4576697415156818e-05,
       "loss": 0.7741,
       "step": 70
     },
     {
       "epoch": 0.4437869822485207,
-      "grad_norm": 0.33684986996070054,
+      "grad_norm": 0.3379144910140547,
       "learning_rate": 1.3635079705638298e-05,
       "loss": 0.7852,
       "step": 75
     },
     {
       "epoch": 0.47337278106508873,
-      "grad_norm": 0.3494701869810424,
+      "grad_norm": 0.353019209730548,
       "learning_rate": 1.2654675551080724e-05,
-      "loss": 0.7666,
+      "loss": 0.7667,
       "step": 80
     },
     {
       "epoch": 0.5029585798816568,
-      "grad_norm": 0.36055423427517397,
+      "grad_norm": 0.36113222294563596,
       "learning_rate": 1.164594590280734e-05,
       "loss": 0.7706,
       "step": 85
     },
     {
       "epoch": 0.5325443786982249,
-      "grad_norm": 0.356808349064,
+      "grad_norm": 0.3620905804084936,
       "learning_rate": 1.0619653946285948e-05,
       "loss": 0.77,
       "step": 90
     },
     {
       "epoch": 0.5621301775147929,
-      "grad_norm": 0.3747307397681612,
+      "grad_norm": 0.3929127481348649,
       "learning_rate": 9.586750257511868e-06,
       "loss": 0.7771,
       "step": 95
     },
     {
       "epoch": 0.591715976331361,
-      "grad_norm": 0.33446919541396347,
+      "grad_norm": 0.33535959778063407,
       "learning_rate": 8.558255959926533e-06,
       "loss": 0.753,
       "step": 100
     },
     {
       "epoch": 0.591715976331361,
-      "eval_loss": 0.7880241870880127,
-      "eval_runtime": 0.7758,
-      "eval_samples_per_second": 164.983,
-      "eval_steps_per_second": 5.156,
+      "eval_loss": 0.7879180908203125,
+      "eval_runtime": 0.7837,
+      "eval_samples_per_second": 163.331,
+      "eval_steps_per_second": 5.104,
       "step": 100
     },
     {
       "epoch": 0.621301775147929,
-      "grad_norm": 0.3341379896313837,
+      "grad_norm": 0.33487543251139984,
       "learning_rate": 7.545145128592009e-06,
       "loss": 0.7569,
       "step": 105
     },
     {
       "epoch": 0.650887573964497,
-      "grad_norm": 0.3449039583059825,
+      "grad_norm": 0.34648632669150914,
       "learning_rate": 6.558227696373617e-06,
       "loss": 0.7681,
       "step": 110
     },
     {
       "epoch": 0.6804733727810651,
-      "grad_norm": 0.30846413126969413,
+      "grad_norm": 0.3082490982179377,
       "learning_rate": 5.608034111526298e-06,
       "loss": 0.7623,
       "step": 115
     },
     {
       "epoch": 0.7100591715976331,
-      "grad_norm": 0.2982369439119382,
+      "grad_norm": 0.2966147844807204,
       "learning_rate": 4.704702977392914e-06,
-      "loss": 0.7513,
+      "loss": 0.7514,
       "step": 120
     },
     {
       "epoch": 0.7396449704142012,
-      "grad_norm": 0.34529781871312787,
+      "grad_norm": 0.3414127530278989,
       "learning_rate": 3.857872873103322e-06,
       "loss": 0.7538,
       "step": 125
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 0.3124440579524454,
+      "grad_norm": 0.31201569821063163,
       "learning_rate": 3.0765795095517026e-06,
       "loss": 0.7555,
       "step": 130
     },
     {
       "epoch": 0.7988165680473372,
-      "grad_norm": 0.32258195329868705,
+      "grad_norm": 0.32174719701784427,
       "learning_rate": 2.369159318001937e-06,
       "loss": 0.7584,
       "step": 135
     },
     {
       "epoch": 0.8284023668639053,
-      "grad_norm": 0.2992110227849431,
+      "grad_norm": 0.2986522028663522,
       "learning_rate": 1.743160500034443e-06,
       "loss": 0.7498,
       "step": 140
     },
     {
       "epoch": 0.8579881656804734,
-      "grad_norm": 0.3122905795517269,
+      "grad_norm": 0.3104418941721675,
       "learning_rate": 1.2052624879351105e-06,
       "loss": 0.7566,
       "step": 145
     },
     {
       "epoch": 0.8875739644970414,
-      "grad_norm": 0.2953430989606355,
+      "grad_norm": 0.29524530948168437,
       "learning_rate": 7.612046748871327e-07,
       "loss": 0.7666,
       "step": 150
     },
     {
       "epoch": 0.9171597633136095,
-      "grad_norm": 0.296575452578378,
+      "grad_norm": 0.29629079484869536,
       "learning_rate": 4.1572517541747294e-07,
       "loss": 0.7613,
       "step": 155
     },
     {
       "epoch": 0.9467455621301775,
-      "grad_norm": 0.28547752704360185,
+      "grad_norm": 0.28577437390363114,
       "learning_rate": 1.7251026952640583e-07,
-      "loss": 0.7607,
+      "loss": 0.7606,
       "step": 160
     },
     {
       "epoch": 0.9763313609467456,
-      "grad_norm": 0.27623679371409765,
+      "grad_norm": 0.2756819768560144,
       "learning_rate": 3.4155069933301535e-08,
       "loss": 0.7445,
       "step": 165
@@ -251,8 +251,8 @@
       "epoch": 1.0,
       "step": 169,
       "total_flos": 76916824473600.0,
-      "train_loss": 0.8026524422436776,
-      "train_runtime": 501.113,
+      "train_loss": 0.8026537260360266,
+      "train_runtime": 501.1144,
       "train_samples_per_second": 43.124,
       "train_steps_per_second": 0.337
     }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4ab3039b64159df3c8e0e6bc3f86b21f6b8ae20d881eca18baa4a6486573b33a
+oid sha256:aa8085e4e6002caed0cedd9845dc90f1d317dbab96d78af19c660dfdc1950ba6
 size 7352