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: 2.2849
+- Loss: 2.2714
 
 ## Model description
 
@@ -57,7 +57,7 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 1.4631        | 0.9961 | 129  | 2.2849          |
+| 1.4571        | 0.9961 | 129  | 2.2714          |
 
 
 ### Framework versions

all_results.json

@@ -1,14 +1,9 @@
 {
     "epoch": 0.9961389961389961,
-    "eval_loss": 2.2848541736602783,
-    "eval_runtime": 1.1208,
-    "eval_samples": 15,
-    "eval_samples_per_second": 10.707,
-    "eval_steps_per_second": 0.892,
     "total_flos": 7.61719099628716e+17,
-    "train_loss": 1.5859730927519096,
-    "train_runtime": 2226.6138,
+    "train_loss": 1.5814437108446462,
+    "train_runtime": 642.9915,
     "train_samples": 111440,
-    "train_samples_per_second": 14.886,
-    "train_steps_per_second": 0.058
+    "train_samples_per_second": 51.55,
+    "train_steps_per_second": 0.201
 }

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 0.9961389961389961,
     "total_flos": 7.61719099628716e+17,
-    "train_loss": 1.5859730927519096,
-    "train_runtime": 2226.6138,
+    "train_loss": 1.5814437108446462,
+    "train_runtime": 642.9915,
     "train_samples": 111440,
-    "train_samples_per_second": 14.886,
-    "train_steps_per_second": 0.058
+    "train_samples_per_second": 51.55,
+    "train_steps_per_second": 0.201
 }

trainer_state.json

@@ -10,202 +10,202 @@
   "log_history": [
     {
       "epoch": 0.007722007722007722,
-      "grad_norm": 0.7566112279891968,
+      "grad_norm": 3.172396421432495,
       "learning_rate": 1.5384615384615387e-05,
-      "loss": 2.1393,
+      "loss": 2.1653,
       "step": 1
     },
     {
       "epoch": 0.03861003861003861,
-      "grad_norm": 1.1990445852279663,
+      "grad_norm": 2.1693484783172607,
       "learning_rate": 7.692307692307693e-05,
-      "loss": 2.1362,
+      "loss": 2.1599,
       "step": 5
     },
     {
       "epoch": 0.07722007722007722,
-      "grad_norm": 0.5208460092544556,
+      "grad_norm": 0.5863114595413208,
       "learning_rate": 0.00015384615384615385,
-      "loss": 2.0814,
+      "loss": 2.0828,
       "step": 10
     },
     {
       "epoch": 0.11583011583011583,
-      "grad_norm": 0.6390016078948975,
+      "grad_norm": 0.6443110108375549,
       "learning_rate": 0.00019985334138511237,
-      "loss": 1.9687,
+      "loss": 1.9643,
       "step": 15
     },
     {
       "epoch": 0.15444015444015444,
-      "grad_norm": 0.6168708801269531,
+      "grad_norm": 0.628454864025116,
       "learning_rate": 0.0001982083682742156,
-      "loss": 1.8165,
+      "loss": 1.813,
       "step": 20
     },
     {
       "epoch": 0.19305019305019305,
-      "grad_norm": 0.5843796133995056,
+      "grad_norm": 0.5933235883712769,
       "learning_rate": 0.00019476531711828027,
-      "loss": 1.6979,
+      "loss": 1.6936,
       "step": 25
     },
     {
       "epoch": 0.23166023166023167,
-      "grad_norm": 0.329062283039093,
+      "grad_norm": 0.31888535618782043,
       "learning_rate": 0.0001895872260758688,
-      "loss": 1.6433,
+      "loss": 1.6382,
       "step": 30
     },
     {
       "epoch": 0.2702702702702703,
-      "grad_norm": 0.35946840047836304,
+      "grad_norm": 0.33554232120513916,
       "learning_rate": 0.00018276889981568906,
-      "loss": 1.5942,
+      "loss": 1.5888,
       "step": 35
     },
     {
       "epoch": 0.3088803088803089,
-      "grad_norm": 0.24955272674560547,
+      "grad_norm": 0.24805951118469238,
       "learning_rate": 0.00017443517375622704,
-      "loss": 1.5573,
+      "loss": 1.5513,
       "step": 40
     },
     {
       "epoch": 0.3474903474903475,
-      "grad_norm": 0.2598620355129242,
+      "grad_norm": 0.2511727213859558,
       "learning_rate": 0.00016473862847818277,
-      "loss": 1.5501,
+      "loss": 1.5439,
       "step": 45
     },
     {
       "epoch": 0.3861003861003861,
-      "grad_norm": 0.21895311772823334,
+      "grad_norm": 0.2083854228258133,
       "learning_rate": 0.00015385679615609042,
-      "loss": 1.5251,
+      "loss": 1.5181,
       "step": 50
     },
     {
       "epoch": 0.4247104247104247,
-      "grad_norm": 0.2305791676044464,
+      "grad_norm": 0.21075661480426788,
       "learning_rate": 0.00014198891015602646,
-      "loss": 1.5191,
+      "loss": 1.5124,
       "step": 55
     },
     {
       "epoch": 0.46332046332046334,
-      "grad_norm": 0.21794967353343964,
+      "grad_norm": 0.20275065302848816,
       "learning_rate": 0.00012935225731039348,
-      "loss": 1.5042,
+      "loss": 1.4972,
       "step": 60
     },
     {
       "epoch": 0.5019305019305019,
-      "grad_norm": 0.22447626292705536,
+      "grad_norm": 0.2260345220565796,
       "learning_rate": 0.0001161781996552765,
-      "loss": 1.4992,
+      "loss": 1.4923,
       "step": 65
     },
     {
       "epoch": 0.5405405405405406,
-      "grad_norm": 0.2490142434835434,
+      "grad_norm": 0.2425995022058487,
       "learning_rate": 0.00010270793846761347,
-      "loss": 1.4967,
+      "loss": 1.4897,
       "step": 70
     },
     {
       "epoch": 0.5791505791505791,
-      "grad_norm": 0.2434530258178711,
+      "grad_norm": 0.23694849014282227,
       "learning_rate": 8.918809815760585e-05,
-      "loss": 1.4842,
+      "loss": 1.4775,
       "step": 75
     },
     {
       "epoch": 0.6177606177606177,
-      "grad_norm": 0.29860490560531616,
+      "grad_norm": 0.26492005586624146,
       "learning_rate": 7.586621087002945e-05,
-      "loss": 1.4808,
+      "loss": 1.474,
       "step": 80
     },
     {
       "epoch": 0.6563706563706564,
-      "grad_norm": 0.2233101725578308,
+      "grad_norm": 0.21545687317848206,
       "learning_rate": 6.298618446600856e-05,
-      "loss": 1.4863,
+      "loss": 1.4795,
       "step": 85
     },
     {
       "epoch": 0.694980694980695,
-      "grad_norm": 0.24076718091964722,
+      "grad_norm": 0.2343147248029709,
       "learning_rate": 5.078383686109926e-05,
-      "loss": 1.4625,
+      "loss": 1.4562,
       "step": 90
     },
     {
       "epoch": 0.7335907335907336,
-      "grad_norm": 0.2131790667772293,
+      "grad_norm": 0.2037033885717392,
       "learning_rate": 3.948257848062351e-05,
-      "loss": 1.4676,
+      "loss": 1.4611,
       "step": 95
     },
     {
       "epoch": 0.7722007722007722,
-      "grad_norm": 0.2535589635372162,
+      "grad_norm": 0.24056674540042877,
       "learning_rate": 2.9289321881345254e-05,
-      "loss": 1.4689,
+      "loss": 1.4629,
       "step": 100
     },
     {
       "epoch": 0.8108108108108109,
-      "grad_norm": 0.2187783420085907,
+      "grad_norm": 0.22205579280853271,
       "learning_rate": 2.0390693429435627e-05,
-      "loss": 1.4613,
+      "loss": 1.4556,
       "step": 105
     },
     {
       "epoch": 0.8494208494208494,
-      "grad_norm": 0.22383730113506317,
+      "grad_norm": 0.2215537130832672,
       "learning_rate": 1.2949616394382802e-05,
-      "loss": 1.4626,
+      "loss": 1.4569,
       "step": 110
     },
     {
       "epoch": 0.888030888030888,
-      "grad_norm": 0.24992400407791138,
+      "grad_norm": 0.2468106597661972,
       "learning_rate": 7.102328018320858e-06,
-      "loss": 1.4698,
+      "loss": 1.4641,
       "step": 115
     },
     {
       "epoch": 0.9266409266409267,
-      "grad_norm": 0.24492764472961426,
+      "grad_norm": 0.23411813378334045,
       "learning_rate": 2.9558851746788517e-06,
-      "loss": 1.4742,
+      "loss": 1.4681,
       "step": 120
     },
     {
       "epoch": 0.9652509652509652,
-      "grad_norm": 0.27442848682403564,
+      "grad_norm": 0.2601025402545929,
       "learning_rate": 5.862042845640403e-07,
-      "loss": 1.4631,
+      "loss": 1.4571,
       "step": 125
     },
     {
       "epoch": 0.9961389961389961,
-      "eval_loss": 2.2848541736602783,
-      "eval_runtime": 1.132,
-      "eval_samples_per_second": 10.601,
-      "eval_steps_per_second": 0.883,
+      "eval_loss": 2.2714059352874756,
+      "eval_runtime": 0.7834,
+      "eval_samples_per_second": 15.317,
+      "eval_steps_per_second": 1.276,
       "step": 129
     },
     {
       "epoch": 0.9961389961389961,
       "step": 129,
       "total_flos": 7.61719099628716e+17,
-      "train_loss": 1.5859730927519096,
-      "train_runtime": 2226.6138,
-      "train_samples_per_second": 14.886,
-      "train_steps_per_second": 0.058
+      "train_loss": 1.5814437108446462,
+      "train_runtime": 642.9915,
+      "train_samples_per_second": 51.55,
+      "train_steps_per_second": 0.201
     }
   ],
   "logging_steps": 5,