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 [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.6480
+- Loss: 1.6455
 
 ## Model description
 
@@ -57,7 +57,7 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 1.1497        | 1.0   | 137  | 1.6480          |
+| 1.1473        | 1.0   | 137  | 1.6455          |
 
 
 ### Framework versions

all_results.json

@@ -1,14 +1,9 @@
 {
     "epoch": 1.0,
-    "eval_loss": 1.6479942798614502,
-    "eval_runtime": 1.1327,
-    "eval_samples": 16,
-    "eval_samples_per_second": 7.945,
-    "eval_steps_per_second": 0.883,
     "total_flos": 8.08957492854784e+17,
-    "train_loss": 1.2461711733880705,
-    "train_runtime": 2338.1563,
+    "train_loss": 1.2450406925521629,
+    "train_runtime": 683.5001,
     "train_samples": 116368,
-    "train_samples_per_second": 14.952,
-    "train_steps_per_second": 0.059
+    "train_samples_per_second": 51.148,
+    "train_steps_per_second": 0.2
 }

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
     "total_flos": 8.08957492854784e+17,
-    "train_loss": 1.2461711733880705,
-    "train_runtime": 2338.1563,
+    "train_loss": 1.2450406925521629,
+    "train_runtime": 683.5001,
     "train_samples": 116368,
-    "train_samples_per_second": 14.952,
-    "train_steps_per_second": 0.059
+    "train_samples_per_second": 51.148,
+    "train_steps_per_second": 0.2
 }

trainer_state.json

@@ -10,216 +10,216 @@
   "log_history": [
     {
       "epoch": 0.0072992700729927005,
-      "grad_norm": 0.43487387895584106,
+      "grad_norm": 0.8994209170341492,
       "learning_rate": 1.4285714285714285e-05,
-      "loss": 1.7554,
+      "loss": 1.7629,
       "step": 1
     },
     {
       "epoch": 0.0364963503649635,
-      "grad_norm": 0.4084155857563019,
+      "grad_norm": 0.7910200953483582,
       "learning_rate": 7.142857142857143e-05,
-      "loss": 1.7248,
+      "loss": 1.7329,
       "step": 5
     },
     {
       "epoch": 0.072992700729927,
-      "grad_norm": 0.4045219123363495,
+      "grad_norm": 0.47016409039497375,
       "learning_rate": 0.00014285714285714287,
-      "loss": 1.6877,
+      "loss": 1.6931,
       "step": 10
     },
     {
       "epoch": 0.10948905109489052,
-      "grad_norm": 0.5852366089820862,
+      "grad_norm": 0.6418158411979675,
       "learning_rate": 0.00019996738360808565,
-      "loss": 1.5983,
+      "loss": 1.5991,
       "step": 15
     },
     {
       "epoch": 0.145985401459854,
-      "grad_norm": 0.5415384769439697,
+      "grad_norm": 0.5754856467247009,
       "learning_rate": 0.00019882804237803488,
-      "loss": 1.4422,
+      "loss": 1.4449,
       "step": 20
     },
     {
       "epoch": 0.18248175182481752,
-      "grad_norm": 0.49430617690086365,
+      "grad_norm": 0.5607455968856812,
       "learning_rate": 0.00019607909582962477,
-      "loss": 1.3509,
+      "loss": 1.3519,
       "step": 25
     },
     {
       "epoch": 0.21897810218978103,
-      "grad_norm": 0.3028700649738312,
+      "grad_norm": 0.31966158747673035,
       "learning_rate": 0.0001917653158603628,
-      "loss": 1.2905,
+      "loss": 1.2901,
       "step": 30
     },
     {
       "epoch": 0.25547445255474455,
-      "grad_norm": 0.21788553893566132,
+      "grad_norm": 0.21750766038894653,
       "learning_rate": 0.00018595696069872013,
-      "loss": 1.2481,
+      "loss": 1.2469,
       "step": 35
     },
     {
       "epoch": 0.291970802919708,
-      "grad_norm": 0.21770533919334412,
+      "grad_norm": 0.21939656138420105,
       "learning_rate": 0.00017874863061334657,
-      "loss": 1.2255,
+      "loss": 1.223,
       "step": 40
     },
     {
       "epoch": 0.3284671532846715,
-      "grad_norm": 0.19103731215000153,
+      "grad_norm": 0.18782173097133636,
       "learning_rate": 0.00017025772716520323,
-      "loss": 1.2213,
+      "loss": 1.2186,
       "step": 45
     },
     {
       "epoch": 0.36496350364963503,
-      "grad_norm": 0.18975041806697845,
+      "grad_norm": 0.19173678755760193,
       "learning_rate": 0.0001606225410966638,
-      "loss": 1.2069,
+      "loss": 1.2035,
       "step": 50
     },
     {
       "epoch": 0.40145985401459855,
-      "grad_norm": 0.21415986120700836,
+      "grad_norm": 0.2181948572397232,
       "learning_rate": 0.00015000000000000001,
-      "loss": 1.1938,
+      "loss": 1.191,
       "step": 55
     },
     {
       "epoch": 0.43795620437956206,
-      "grad_norm": 0.20551565289497375,
+      "grad_norm": 0.20430037379264832,
       "learning_rate": 0.0001385631124488136,
-      "loss": 1.1837,
+      "loss": 1.1811,
       "step": 60
     },
     {
       "epoch": 0.4744525547445255,
-      "grad_norm": 0.20549848675727844,
+      "grad_norm": 0.19574333727359772,
       "learning_rate": 0.0001264981502196662,
-      "loss": 1.1724,
+      "loss": 1.1705,
       "step": 65
     },
     {
       "epoch": 0.5109489051094891,
-      "grad_norm": 0.19810882210731506,
+      "grad_norm": 0.1992381513118744,
       "learning_rate": 0.00011400161449686293,
-      "loss": 1.1713,
+      "loss": 1.1695,
       "step": 70
     },
     {
       "epoch": 0.5474452554744526,
-      "grad_norm": 0.18478353321552277,
+      "grad_norm": 0.18511532247066498,
       "learning_rate": 0.00010127703547159739,
-      "loss": 1.1571,
+      "loss": 1.1559,
       "step": 75
     },
     {
       "epoch": 0.583941605839416,
-      "grad_norm": 0.19806450605392456,
+      "grad_norm": 0.19379611313343048,
       "learning_rate": 8.853165746015997e-05,
-      "loss": 1.1539,
+      "loss": 1.1526,
       "step": 80
     },
     {
       "epoch": 0.6204379562043796,
-      "grad_norm": 0.20397868752479553,
+      "grad_norm": 0.19651520252227783,
       "learning_rate": 7.597306353045393e-05,
-      "loss": 1.1457,
+      "loss": 1.1439,
       "step": 85
     },
     {
       "epoch": 0.656934306569343,
-      "grad_norm": 0.18514488637447357,
+      "grad_norm": 0.18215572834014893,
       "learning_rate": 6.380579461128819e-05,
-      "loss": 1.1613,
+      "loss": 1.1592,
       "step": 90
     },
     {
       "epoch": 0.6934306569343066,
-      "grad_norm": 0.18412043154239655,
+      "grad_norm": 0.19182553887367249,
       "learning_rate": 5.222801814877369e-05,
-      "loss": 1.1449,
+      "loss": 1.1425,
       "step": 95
     },
     {
       "epoch": 0.7299270072992701,
-      "grad_norm": 0.19028052687644958,
+      "grad_norm": 0.19180874526500702,
       "learning_rate": 4.142830056718052e-05,
-      "loss": 1.1511,
+      "loss": 1.1488,
       "step": 100
     },
     {
       "epoch": 0.7664233576642335,
-      "grad_norm": 0.1979902684688568,
+      "grad_norm": 0.19942091405391693,
       "learning_rate": 3.158253610095697e-05,
-      "loss": 1.1457,
+      "loss": 1.1437,
       "step": 105
     },
     {
       "epoch": 0.8029197080291971,
-      "grad_norm": 0.19014957547187805,
+      "grad_norm": 0.19039174914360046,
       "learning_rate": 2.2851082017805703e-05,
-      "loss": 1.1423,
+      "loss": 1.1397,
       "step": 110
     },
     {
       "epoch": 0.8394160583941606,
-      "grad_norm": 0.186836376786232,
+      "grad_norm": 0.18815937638282776,
       "learning_rate": 1.5376146891235598e-05,
-      "loss": 1.1477,
+      "loss": 1.1451,
       "step": 115
     },
     {
       "epoch": 0.8759124087591241,
-      "grad_norm": 0.20176666975021362,
+      "grad_norm": 0.19718731939792633,
       "learning_rate": 9.279474459608805e-06,
-      "loss": 1.153,
+      "loss": 1.1502,
       "step": 120
     },
     {
       "epoch": 0.9124087591240876,
-      "grad_norm": 0.19549883902072906,
+      "grad_norm": 0.19630739092826843,
       "learning_rate": 4.660360794506946e-06,
-      "loss": 1.1444,
+      "loss": 1.1419,
       "step": 125
     },
     {
       "epoch": 0.948905109489051,
-      "grad_norm": 0.17397421598434448,
+      "grad_norm": 0.1804933249950409,
       "learning_rate": 1.5940370726542863e-06,
-      "loss": 1.1508,
+      "loss": 1.1483,
       "step": 130
     },
     {
       "epoch": 0.9854014598540146,
-      "grad_norm": 0.1897565871477127,
+      "grad_norm": 0.1939323991537094,
       "learning_rate": 1.3044429107700318e-07,
-      "loss": 1.1497,
+      "loss": 1.1473,
       "step": 135
     },
     {
       "epoch": 1.0,
-      "eval_loss": 1.6479942798614502,
-      "eval_runtime": 1.1173,
-      "eval_samples_per_second": 8.055,
-      "eval_steps_per_second": 0.895,
+      "eval_loss": 1.645493507385254,
+      "eval_runtime": 0.792,
+      "eval_samples_per_second": 11.363,
+      "eval_steps_per_second": 1.263,
       "step": 137
     },
     {
       "epoch": 1.0,
       "step": 137,
       "total_flos": 8.08957492854784e+17,
-      "train_loss": 1.2461711733880705,
-      "train_runtime": 2338.1563,
-      "train_samples_per_second": 14.952,
-      "train_steps_per_second": 0.059
+      "train_loss": 1.2450406925521629,
+      "train_runtime": 683.5001,
+      "train_samples_per_second": 51.148,
+      "train_steps_per_second": 0.2
     }
   ],
   "logging_steps": 5,