Model save

README.md

@@ -20,7 +20,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the generator dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.3992
+- Loss: 1.4024
 
 ## Model description
 
@@ -57,7 +57,7 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 0.8942        | 1.0   | 185  | 1.3992          |
+| 0.893         | 1.0   | 185  | 1.4024          |
 
 
 ### Framework versions

all_results.json

@@ -1,14 +1,9 @@
 {
     "epoch": 1.0,
-    "eval_loss": 1.3991833925247192,
-    "eval_runtime": 1.4894,
-    "eval_samples": 16,
-    "eval_samples_per_second": 7.385,
-    "eval_steps_per_second": 0.671,
     "total_flos": 9.060485625492275e+17,
-    "train_loss": 0.9615903693276483,
-    "train_runtime": 3138.0854,
+    "train_loss": 0.9603648733448338,
+    "train_runtime": 724.3175,
     "train_samples": 116368,
-    "train_samples_per_second": 13.203,
-    "train_steps_per_second": 0.059
+    "train_samples_per_second": 57.2,
+    "train_steps_per_second": 0.255
 }

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
     "total_flos": 9.060485625492275e+17,
-    "train_loss": 0.9615903693276483,
-    "train_runtime": 3138.0854,
+    "train_loss": 0.9603648733448338,
+    "train_runtime": 724.3175,
     "train_samples": 116368,
-    "train_samples_per_second": 13.203,
-    "train_steps_per_second": 0.059
+    "train_samples_per_second": 57.2,
+    "train_steps_per_second": 0.255
 }

trainer_state.json

@@ -10,286 +10,286 @@
   "log_history": [
     {
       "epoch": 0.005405405405405406,
-      "grad_norm": 2.6925952434539795,
+      "grad_norm": 2.6678006649017334,
       "learning_rate": 1.0526315789473684e-05,
-      "loss": 1.4517,
+      "loss": 1.4457,
       "step": 1
     },
     {
       "epoch": 0.02702702702702703,
-      "grad_norm": 2.4539754390716553,
+      "grad_norm": 2.4659605026245117,
       "learning_rate": 5.2631578947368424e-05,
-      "loss": 1.4341,
+      "loss": 1.4311,
       "step": 5
     },
     {
       "epoch": 0.05405405405405406,
-      "grad_norm": 2.1194558143615723,
+      "grad_norm": 2.074678659439087,
       "learning_rate": 0.00010526315789473685,
-      "loss": 1.3761,
+      "loss": 1.376,
       "step": 10
     },
     {
       "epoch": 0.08108108108108109,
-      "grad_norm": 2.0553150177001953,
+      "grad_norm": 1.9776719808578491,
       "learning_rate": 0.00015789473684210527,
-      "loss": 1.2687,
+      "loss": 1.2676,
       "step": 15
     },
     {
       "epoch": 0.10810810810810811,
-      "grad_norm": 1.8468493223190308,
+      "grad_norm": 1.8173789978027344,
       "learning_rate": 0.00019998209226697376,
-      "loss": 1.1537,
+      "loss": 1.1517,
       "step": 20
     },
     {
       "epoch": 0.13513513513513514,
-      "grad_norm": 1.0270365476608276,
+      "grad_norm": 1.0727437734603882,
       "learning_rate": 0.0001993559947963185,
-      "loss": 1.0561,
+      "loss": 1.0512,
       "step": 25
     },
     {
       "epoch": 0.16216216216216217,
-      "grad_norm": 0.9126940369606018,
+      "grad_norm": 0.8763325214385986,
       "learning_rate": 0.00019784091409455728,
-      "loss": 0.9985,
+      "loss": 0.9971,
       "step": 30
     },
     {
       "epoch": 0.1891891891891892,
-      "grad_norm": 0.6563550233840942,
+      "grad_norm": 0.6051992774009705,
       "learning_rate": 0.0001954504062771555,
-      "loss": 0.9701,
+      "loss": 0.9685,
       "step": 35
     },
     {
       "epoch": 0.21621621621621623,
-      "grad_norm": 0.6633013486862183,
+      "grad_norm": 0.6363440752029419,
       "learning_rate": 0.00019220586030376134,
-      "loss": 0.9619,
+      "loss": 0.9609,
       "step": 40
     },
     {
       "epoch": 0.24324324324324326,
-      "grad_norm": 0.7960276007652283,
+      "grad_norm": 0.7076846957206726,
       "learning_rate": 0.00018813630660146488,
-      "loss": 0.9392,
+      "loss": 0.938,
       "step": 45
     },
     {
       "epoch": 0.2702702702702703,
-      "grad_norm": 0.6284172534942627,
+      "grad_norm": 0.5835558772087097,
       "learning_rate": 0.00018327815731637612,
-      "loss": 0.9387,
+      "loss": 0.9376,
       "step": 50
     },
     {
       "epoch": 0.2972972972972973,
-      "grad_norm": 0.6241137981414795,
+      "grad_norm": 0.6069886088371277,
       "learning_rate": 0.00017767488051760857,
-      "loss": 0.9316,
+      "loss": 0.9303,
       "step": 55
     },
     {
       "epoch": 0.32432432432432434,
-      "grad_norm": 0.7423997521400452,
+      "grad_norm": 0.7000340819358826,
       "learning_rate": 0.0001713766112687139,
-      "loss": 0.9287,
+      "loss": 0.9276,
       "step": 60
     },
     {
       "epoch": 0.35135135135135137,
-      "grad_norm": 0.6817660331726074,
+      "grad_norm": 0.6510019898414612,
       "learning_rate": 0.0001644397030464877,
-      "loss": 0.9293,
+      "loss": 0.9282,
       "step": 65
     },
     {
       "epoch": 0.3783783783783784,
-      "grad_norm": 0.6054185628890991,
+      "grad_norm": 0.6287819147109985,
       "learning_rate": 0.00015692622352080662,
-      "loss": 0.9134,
+      "loss": 0.9125,
       "step": 70
     },
     {
       "epoch": 0.40540540540540543,
-      "grad_norm": 0.6050704121589661,
+      "grad_norm": 0.6086368560791016,
       "learning_rate": 0.00014890339920698334,
-      "loss": 0.93,
+      "loss": 0.9291,
       "step": 75
     },
     {
       "epoch": 0.43243243243243246,
-      "grad_norm": 0.589340090751648,
+      "grad_norm": 0.5720112919807434,
       "learning_rate": 0.0001404430139595877,
-      "loss": 0.9163,
+      "loss": 0.9152,
       "step": 80
     },
     {
       "epoch": 0.4594594594594595,
-      "grad_norm": 0.6608156561851501,
+      "grad_norm": 0.6146399974822998,
       "learning_rate": 0.0001316207666896824,
-      "loss": 0.9211,
+      "loss": 0.9201,
       "step": 85
     },
     {
       "epoch": 0.4864864864864865,
-      "grad_norm": 0.5910987854003906,
+      "grad_norm": 0.5874430537223816,
       "learning_rate": 0.00012251559405226941,
-      "loss": 0.9081,
+      "loss": 0.9071,
       "step": 90
     },
     {
       "epoch": 0.5135135135135135,
-      "grad_norm": 0.6293689608573914,
+      "grad_norm": 0.609653115272522,
       "learning_rate": 0.00011320896416417026,
-      "loss": 0.9123,
+      "loss": 0.9111,
       "step": 95
     },
     {
       "epoch": 0.5405405405405406,
-      "grad_norm": 0.5938843488693237,
+      "grad_norm": 0.5806834101676941,
       "learning_rate": 0.00010378414767176705,
-      "loss": 0.902,
+      "loss": 0.9008,
       "step": 100
     },
     {
       "epoch": 0.5675675675675675,
-      "grad_norm": 0.6008319854736328,
+      "grad_norm": 0.5760082602500916,
       "learning_rate": 9.432547269069261e-05,
-      "loss": 0.9061,
+      "loss": 0.9053,
       "step": 105
     },
     {
       "epoch": 0.5945945945945946,
-      "grad_norm": 0.6468297839164734,
+      "grad_norm": 0.6656131148338318,
       "learning_rate": 8.491757028386263e-05,
-      "loss": 0.9038,
+      "loss": 0.9029,
       "step": 110
     },
     {
       "epoch": 0.6216216216216216,
-      "grad_norm": 0.6693244576454163,
+      "grad_norm": 0.6559913754463196,
       "learning_rate": 7.564461722890081e-05,
-      "loss": 0.911,
+      "loss": 0.9103,
       "step": 115
     },
     {
       "epoch": 0.6486486486486487,
-      "grad_norm": 0.590085506439209,
+      "grad_norm": 0.5813584327697754,
       "learning_rate": 6.658958285026102e-05,
-      "loss": 0.9041,
+      "loss": 0.9033,
       "step": 120
     },
     {
       "epoch": 0.6756756756756757,
-      "grad_norm": 0.5497063398361206,
+      "grad_norm": 0.538943350315094,
       "learning_rate": 5.7833486654981606e-05,
-      "loss": 0.9066,
+      "loss": 0.9059,
       "step": 125
     },
     {
       "epoch": 0.7027027027027027,
-      "grad_norm": 0.5866259932518005,
+      "grad_norm": 0.5927494764328003,
       "learning_rate": 4.945467341434195e-05,
-      "loss": 0.904,
+      "loss": 0.9031,
       "step": 130
     },
     {
       "epoch": 0.7297297297297297,
-      "grad_norm": 0.6213558316230774,
+      "grad_norm": 0.6069759726524353,
       "learning_rate": 4.152811217759529e-05,
-      "loss": 0.9036,
+      "loss": 0.9027,
       "step": 135
     },
     {
       "epoch": 0.7567567567567568,
-      "grad_norm": 0.5839004516601562,
+      "grad_norm": 0.5718995332717896,
       "learning_rate": 3.4124725489820645e-05,
-      "loss": 0.8903,
+      "loss": 0.8891,
       "step": 140
     },
     {
       "epoch": 0.7837837837837838,
-      "grad_norm": 0.6068409085273743,
+      "grad_norm": 0.6052721738815308,
       "learning_rate": 2.7310754815685624e-05,
-      "loss": 0.8978,
+      "loss": 0.8972,
       "step": 145
     },
     {
       "epoch": 0.8108108108108109,
-      "grad_norm": 0.6482556462287903,
+      "grad_norm": 0.6083750128746033,
       "learning_rate": 2.1147167846963422e-05,
-      "loss": 0.8958,
+      "loss": 0.8948,
       "step": 150
     },
     {
       "epoch": 0.8378378378378378,
-      "grad_norm": 0.5723783373832703,
+      "grad_norm": 0.5737211108207703,
       "learning_rate": 1.5689112996891576e-05,
-      "loss": 0.9028,
+      "loss": 0.9016,
       "step": 155
     },
     {
       "epoch": 0.8648648648648649,
-      "grad_norm": 0.655846357345581,
+      "grad_norm": 0.6462253332138062,
       "learning_rate": 1.0985425962260343e-05,
-      "loss": 0.8999,
+      "loss": 0.8989,
       "step": 160
     },
     {
       "epoch": 0.8918918918918919,
-      "grad_norm": 0.6444997787475586,
+      "grad_norm": 0.6070159077644348,
       "learning_rate": 7.078192768243486e-06,
-      "loss": 0.8923,
+      "loss": 0.8913,
       "step": 165
     },
     {
       "epoch": 0.918918918918919,
-      "grad_norm": 0.5838979482650757,
+      "grad_norm": 0.5784199833869934,
       "learning_rate": 4.002373205607723e-06,
-      "loss": 0.8973,
+      "loss": 0.8962,
       "step": 170
     },
     {
       "epoch": 0.9459459459459459,
-      "grad_norm": 0.6335191130638123,
+      "grad_norm": 0.6253275871276855,
       "learning_rate": 1.7854880295797405e-06,
-      "loss": 0.8826,
+      "loss": 0.8815,
       "step": 175
     },
     {
       "epoch": 0.972972972972973,
-      "grad_norm": 0.5943057537078857,
+      "grad_norm": 0.5931078195571899,
       "learning_rate": 4.4737271914411236e-07,
-      "loss": 0.8931,
+      "loss": 0.8919,
       "step": 180
     },
     {
       "epoch": 1.0,
-      "grad_norm": 0.600213885307312,
+      "grad_norm": 0.6198851466178894,
       "learning_rate": 0.0,
-      "loss": 0.8942,
+      "loss": 0.893,
       "step": 185
     },
     {
       "epoch": 1.0,
-      "eval_loss": 1.3991833925247192,
-      "eval_runtime": 1.4887,
-      "eval_samples_per_second": 7.389,
-      "eval_steps_per_second": 0.672,
+      "eval_loss": 1.4024296998977661,
+      "eval_runtime": 0.674,
+      "eval_samples_per_second": 16.321,
+      "eval_steps_per_second": 1.484,
       "step": 185
     },
     {
       "epoch": 1.0,
       "step": 185,
       "total_flos": 9.060485625492275e+17,
-      "train_loss": 0.9615903693276483,
-      "train_runtime": 3138.0854,
-      "train_samples_per_second": 13.203,
-      "train_steps_per_second": 0.059
+      "train_loss": 0.9603648733448338,
+      "train_runtime": 724.3175,
+      "train_samples_per_second": 57.2,
+      "train_steps_per_second": 0.255
     }
   ],
   "logging_steps": 5,