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/llmsft/huggingface/runs/t99f8ln3) 
+[<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/llmsft/huggingface/runs/m76fwcrh) 
 
 
 This model was trained with SFT.

all_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
     "total_flos": 76916824473600.0,
-    "train_loss": 0.8026211684977514,
-    "train_runtime": 531.3382,
+    "train_loss": 0.8026510822702442,
+    "train_runtime": 538.1489,
     "train_samples": 16610,
-    "train_samples_per_second": 40.671,
-    "train_steps_per_second": 0.318
+    "train_samples_per_second": 40.156,
+    "train_steps_per_second": 0.314
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e912459075df81904a762019ffac7e46dd397ff0946ecf62ea965cfa6aedc0db
+oid sha256:b54efcd4779102681181b3d5e87aea3aa0b63ec7af455f56462d3e5abce86d7d
 size 3087467144

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
     "total_flos": 76916824473600.0,
-    "train_loss": 0.8026211684977514,
-    "train_runtime": 531.3382,
+    "train_loss": 0.8026510822702442,
+    "train_runtime": 538.1489,
     "train_samples": 16610,
-    "train_samples_per_second": 40.671,
-    "train_steps_per_second": 0.318
+    "train_samples_per_second": 40.156,
+    "train_steps_per_second": 0.314
 }

trainer_state.json

@@ -10,239 +10,239 @@
   "log_history": [
     {
       "epoch": 0.029585798816568046,
-      "grad_norm": 2.344656673971504,
+      "grad_norm": 2.34431495094475,
       "learning_rate": 5.882352941176471e-06,
       "loss": 1.0993,
       "step": 5
     },
     {
       "epoch": 0.05917159763313609,
-      "grad_norm": 1.6349323509871891,
+      "grad_norm": 1.6342466055788052,
       "learning_rate": 1.1764705882352942e-05,
-      "loss": 1.0403,
+      "loss": 1.0402,
       "step": 10
     },
     {
       "epoch": 0.08875739644970414,
-      "grad_norm": 0.8887788496534335,
+      "grad_norm": 0.8873124675504254,
       "learning_rate": 1.7647058823529414e-05,
-      "loss": 0.9534,
+      "loss": 0.9533,
       "step": 15
     },
     {
       "epoch": 0.11834319526627218,
-      "grad_norm": 0.6061805825121692,
+      "grad_norm": 0.6061384548577425,
       "learning_rate": 1.9980782984658682e-05,
       "loss": 0.8923,
       "step": 20
     },
     {
       "epoch": 0.14792899408284024,
-      "grad_norm": 0.539131414229794,
+      "grad_norm": 0.5394301402760198,
       "learning_rate": 1.9863613034027224e-05,
       "loss": 0.8552,
       "step": 25
     },
     {
       "epoch": 0.17751479289940827,
-      "grad_norm": 0.42631321137779,
+      "grad_norm": 0.4263004499359231,
       "learning_rate": 1.9641197940012136e-05,
       "loss": 0.8283,
       "step": 30
     },
     {
       "epoch": 0.20710059171597633,
-      "grad_norm": 0.37715489950270836,
+      "grad_norm": 0.3765938385799202,
       "learning_rate": 1.9315910880512792e-05,
-      "loss": 0.8229,
+      "loss": 0.823,
       "step": 35
     },
     {
       "epoch": 0.23668639053254437,
-      "grad_norm": 0.3979093767540078,
+      "grad_norm": 0.3994617015326158,
       "learning_rate": 1.8891222681391853e-05,
-      "loss": 0.8224,
+      "loss": 0.8225,
       "step": 40
     },
     {
       "epoch": 0.26627218934911245,
-      "grad_norm": 0.34245201479374676,
+      "grad_norm": 0.3435049422799343,
       "learning_rate": 1.8371664782625287e-05,
-      "loss": 0.8073,
+      "loss": 0.8074,
       "step": 45
     },
     {
       "epoch": 0.2958579881656805,
-      "grad_norm": 0.3341648838115499,
+      "grad_norm": 0.3345944062547487,
       "learning_rate": 1.7762780887657576e-05,
-      "loss": 0.7977,
+      "loss": 0.7978,
       "step": 50
     },
     {
       "epoch": 0.3254437869822485,
-      "grad_norm": 0.3670051272036132,
+      "grad_norm": 0.36715317735714187,
       "learning_rate": 1.7071067811865477e-05,
-      "loss": 0.7876,
+      "loss": 0.7877,
       "step": 55
     },
     {
       "epoch": 0.35502958579881655,
-      "grad_norm": 0.37158220090572625,
+      "grad_norm": 0.4101583924207969,
       "learning_rate": 1.6303906161279554e-05,
       "loss": 0.7981,
       "step": 60
     },
     {
       "epoch": 0.38461538461538464,
-      "grad_norm": 0.3417059582199807,
+      "grad_norm": 0.3429472808013907,
       "learning_rate": 1.5469481581224274e-05,
-      "loss": 0.7721,
+      "loss": 0.7722,
       "step": 65
     },
     {
       "epoch": 0.41420118343195267,
-      "grad_norm": 0.3465650208668149,
+      "grad_norm": 0.34459613547301815,
       "learning_rate": 1.4576697415156818e-05,
-      "loss": 0.774,
+      "loss": 0.7741,
       "step": 70
     },
     {
       "epoch": 0.4437869822485207,
-      "grad_norm": 0.33758867730245407,
+      "grad_norm": 0.3366816158560283,
       "learning_rate": 1.3635079705638298e-05,
       "loss": 0.7852,
       "step": 75
     },
     {
       "epoch": 0.47337278106508873,
-      "grad_norm": 0.3524456214278009,
+      "grad_norm": 0.3498335125743547,
       "learning_rate": 1.2654675551080724e-05,
       "loss": 0.7666,
       "step": 80
     },
     {
       "epoch": 0.5029585798816568,
-      "grad_norm": 0.3614928061123476,
+      "grad_norm": 0.36240463922758526,
       "learning_rate": 1.164594590280734e-05,
       "loss": 0.7706,
       "step": 85
     },
     {
       "epoch": 0.5325443786982249,
-      "grad_norm": 0.35946654149234436,
+      "grad_norm": 0.3598709812235145,
       "learning_rate": 1.0619653946285948e-05,
       "loss": 0.77,
       "step": 90
     },
     {
       "epoch": 0.5621301775147929,
-      "grad_norm": 0.3735061631999573,
+      "grad_norm": 0.37079937891628517,
       "learning_rate": 9.586750257511868e-06,
       "loss": 0.7771,
       "step": 95
     },
     {
       "epoch": 0.591715976331361,
-      "grad_norm": 0.33319398964601016,
+      "grad_norm": 0.33426442962761077,
       "learning_rate": 8.558255959926533e-06,
       "loss": 0.753,
       "step": 100
     },
     {
       "epoch": 0.591715976331361,
-      "eval_loss": 0.787921130657196,
-      "eval_runtime": 0.8101,
-      "eval_samples_per_second": 157.996,
-      "eval_steps_per_second": 4.937,
+      "eval_loss": 0.7879854440689087,
+      "eval_runtime": 0.8163,
+      "eval_samples_per_second": 156.812,
+      "eval_steps_per_second": 4.9,
       "step": 100
     },
     {
       "epoch": 0.621301775147929,
-      "grad_norm": 0.33520661745738783,
+      "grad_norm": 0.3345582245144445,
       "learning_rate": 7.545145128592009e-06,
       "loss": 0.7569,
       "step": 105
     },
     {
       "epoch": 0.650887573964497,
-      "grad_norm": 0.3448413371244227,
+      "grad_norm": 0.34432279642970004,
       "learning_rate": 6.558227696373617e-06,
-      "loss": 0.768,
+      "loss": 0.7681,
       "step": 110
     },
     {
       "epoch": 0.6804733727810651,
-      "grad_norm": 0.3079769094135903,
+      "grad_norm": 0.3086411269426158,
       "learning_rate": 5.608034111526298e-06,
-      "loss": 0.7622,
+      "loss": 0.7623,
       "step": 115
     },
     {
       "epoch": 0.7100591715976331,
-      "grad_norm": 0.2972264032686521,
+      "grad_norm": 0.2978345987483447,
       "learning_rate": 4.704702977392914e-06,
       "loss": 0.7513,
       "step": 120
     },
     {
       "epoch": 0.7396449704142012,
-      "grad_norm": 0.3441093510555947,
+      "grad_norm": 0.34605624289781844,
       "learning_rate": 3.857872873103322e-06,
-      "loss": 0.7537,
+      "loss": 0.7538,
       "step": 125
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 0.3112946197453334,
+      "grad_norm": 0.3123769318472645,
       "learning_rate": 3.0765795095517026e-06,
       "loss": 0.7555,
       "step": 130
     },
     {
       "epoch": 0.7988165680473372,
-      "grad_norm": 0.32214494904677926,
+      "grad_norm": 0.32257906199571956,
       "learning_rate": 2.369159318001937e-06,
       "loss": 0.7584,
       "step": 135
     },
     {
       "epoch": 0.8284023668639053,
-      "grad_norm": 0.2988407251124208,
+      "grad_norm": 0.29735684576283516,
       "learning_rate": 1.743160500034443e-06,
-      "loss": 0.7497,
+      "loss": 0.7498,
       "step": 140
     },
     {
       "epoch": 0.8579881656804734,
-      "grad_norm": 0.3110738204304494,
+      "grad_norm": 0.3124434574150907,
       "learning_rate": 1.2052624879351105e-06,
-      "loss": 0.7565,
+      "loss": 0.7566,
       "step": 145
     },
     {
       "epoch": 0.8875739644970414,
-      "grad_norm": 0.2946131078163015,
+      "grad_norm": 0.29471192094816323,
       "learning_rate": 7.612046748871327e-07,
       "loss": 0.7665,
       "step": 150
     },
     {
       "epoch": 0.9171597633136095,
-      "grad_norm": 0.2967665065102847,
+      "grad_norm": 0.2975392552888192,
       "learning_rate": 4.1572517541747294e-07,
       "loss": 0.7613,
       "step": 155
     },
     {
       "epoch": 0.9467455621301775,
-      "grad_norm": 0.2863892215513516,
+      "grad_norm": 0.28636682471900105,
       "learning_rate": 1.7251026952640583e-07,
       "loss": 0.7606,
       "step": 160
     },
     {
       "epoch": 0.9763313609467456,
-      "grad_norm": 0.2766071193489955,
+      "grad_norm": 0.2757165715925127,
       "learning_rate": 3.4155069933301535e-08,
       "loss": 0.7445,
       "step": 165
@@ -251,10 +251,10 @@
       "epoch": 1.0,
       "step": 169,
       "total_flos": 76916824473600.0,
-      "train_loss": 0.8026211684977514,
-      "train_runtime": 531.3382,
-      "train_samples_per_second": 40.671,
-      "train_steps_per_second": 0.318
+      "train_loss": 0.8026510822702442,
+      "train_runtime": 538.1489,
+      "train_samples_per_second": 40.156,
+      "train_steps_per_second": 0.314
     }
   ],
   "logging_steps": 5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76c23d7c6df460377176ca92bed5fc8709787fee3b2599ddf3835757514cf7f3
+oid sha256:8e580319d2f67fbfb9f03838eb08f39572c9c66ce022e428c05391c3eb5bfa89
 size 7352