MHGanainy/16-clusters-balanced-10

README.md

@@ -15,6 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
 # 16-clusters-balanced-10
 
 This model is a fine-tuned version of [openai-community/gpt2-xl](https://huggingface.co/openai-community/gpt2-xl) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.8892
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 1.0,
-    "eval_loss": 1.887420654296875,
-    "eval_runtime": 30.0268,
-    "eval_samples_per_second": 13.788,
-    "eval_steps_per_second": 1.732,
-    "perplexity": 6.602317041368738,
+    "eval_loss": 1.8891628980636597,
+    "eval_runtime": 30.0732,
+    "eval_samples_per_second": 13.766,
+    "eval_steps_per_second": 1.729,
+    "perplexity": 6.613829913282871,
     "total_flos": 7.32108351012864e+16,
-    "train_loss": 2.103881771554424,
-    "train_runtime": 1250.9785,
-    "train_samples_per_second": 6.441,
-    "train_steps_per_second": 3.221
+    "train_loss": 2.105005581936963,
+    "train_runtime": 1251.4031,
+    "train_samples_per_second": 6.438,
+    "train_steps_per_second": 3.22
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_loss": 1.887420654296875,
-    "eval_runtime": 30.0268,
-    "eval_samples_per_second": 13.788,
-    "eval_steps_per_second": 1.732,
-    "perplexity": 6.602317041368738
+    "eval_loss": 1.8891628980636597,
+    "eval_runtime": 30.0732,
+    "eval_samples_per_second": 13.766,
+    "eval_steps_per_second": 1.729,
+    "perplexity": 6.613829913282871
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
     "total_flos": 7.32108351012864e+16,
-    "train_loss": 2.103881771554424,
-    "train_runtime": 1250.9785,
-    "train_samples_per_second": 6.441,
-    "train_steps_per_second": 3.221
+    "train_loss": 2.105005581936963,
+    "train_runtime": 1251.4031,
+    "train_samples_per_second": 6.438,
+    "train_steps_per_second": 3.22
 }

trainer_state.json

@@ -10,292 +10,292 @@
   "log_history": [
     {
       "epoch": 0.02482005460412013,
-      "grad_norm": 0.13781653344631195,
+      "grad_norm": 0.1369238644838333,
       "learning_rate": 6.666666666666667e-06,
       "loss": 2.4042,
       "step": 100
     },
     {
       "epoch": 0.04964010920824026,
-      "grad_norm": 0.16998766362667084,
+      "grad_norm": 0.17341101169586182,
       "learning_rate": 1.3333333333333333e-05,
-      "loss": 2.3579,
+      "loss": 2.3581,
       "step": 200
     },
     {
       "epoch": 0.07446016381236038,
-      "grad_norm": 0.2991155982017517,
+      "grad_norm": 0.30298689007759094,
       "learning_rate": 2e-05,
-      "loss": 2.2975,
+      "loss": 2.2979,
       "step": 300
     },
     {
       "epoch": 0.09928021841648052,
-      "grad_norm": 0.4158655107021332,
+      "grad_norm": 0.4181392788887024,
       "learning_rate": 1.9964532702725803e-05,
-      "loss": 2.2728,
+      "loss": 2.2736,
       "step": 400
     },
     {
       "epoch": 0.12410027302060064,
-      "grad_norm": 0.4808669984340668,
+      "grad_norm": 0.4833754301071167,
       "learning_rate": 1.9858382396738395e-05,
-      "loss": 2.2344,
+      "loss": 2.2352,
       "step": 500
     },
     {
       "epoch": 0.14892032762472077,
-      "grad_norm": 0.5469107031822205,
+      "grad_norm": 0.5508949756622314,
       "learning_rate": 1.9682302054929414e-05,
-      "loss": 2.194,
+      "loss": 2.1951,
       "step": 600
     },
     {
       "epoch": 0.17374038222884092,
-      "grad_norm": 0.5828746557235718,
+      "grad_norm": 0.5856565833091736,
       "learning_rate": 1.943754069606428e-05,
-      "loss": 2.1653,
+      "loss": 2.1662,
       "step": 700
     },
     {
       "epoch": 0.19856043683296104,
-      "grad_norm": 0.5619407892227173,
+      "grad_norm": 0.5611233115196228,
       "learning_rate": 1.9125834524918215e-05,
-      "loss": 2.1805,
+      "loss": 2.1815,
       "step": 800
     },
     {
       "epoch": 0.22338049143708116,
-      "grad_norm": 0.6732242107391357,
+      "grad_norm": 0.6802138090133667,
       "learning_rate": 1.8749394616578068e-05,
-      "loss": 2.1666,
+      "loss": 2.1675,
       "step": 900
     },
     {
       "epoch": 0.2482005460412013,
-      "grad_norm": 0.6464433670043945,
+      "grad_norm": 0.6513592004776001,
       "learning_rate": 1.8310891232270827e-05,
-      "loss": 2.1392,
+      "loss": 2.1402,
       "step": 1000
     },
     {
       "epoch": 0.2730206006453214,
-      "grad_norm": 0.6885677576065063,
+      "grad_norm": 0.6889598369598389,
       "learning_rate": 1.781343487797389e-05,
-      "loss": 2.1322,
+      "loss": 2.1334,
       "step": 1100
     },
     {
       "epoch": 0.29784065524944153,
-      "grad_norm": 0.7929017543792725,
+      "grad_norm": 0.7928256988525391,
       "learning_rate": 1.7260554240167017e-05,
-      "loss": 2.1284,
+      "loss": 2.1295,
       "step": 1200
     },
     {
       "epoch": 0.32266070985356166,
-      "grad_norm": 0.7100329399108887,
+      "grad_norm": 0.7162489295005798,
       "learning_rate": 1.665617115523785e-05,
-      "loss": 2.122,
+      "loss": 2.1232,
       "step": 1300
     },
     {
       "epoch": 0.34748076445768183,
-      "grad_norm": 0.7074483633041382,
+      "grad_norm": 0.7136086225509644,
       "learning_rate": 1.6004572790094535e-05,
-      "loss": 2.1137,
+      "loss": 2.1148,
       "step": 1400
     },
     {
       "epoch": 0.37230081906180196,
-      "grad_norm": 0.769305408000946,
+      "grad_norm": 0.7688263654708862,
       "learning_rate": 1.531038123132105e-05,
-      "loss": 2.0861,
+      "loss": 2.0873,
       "step": 1500
     },
     {
       "epoch": 0.3971208736659221,
-      "grad_norm": 0.7742049098014832,
+      "grad_norm": 0.772521436214447,
       "learning_rate": 1.4578520698593441e-05,
-      "loss": 2.1159,
+      "loss": 2.117,
       "step": 1600
     },
     {
       "epoch": 0.4219409282700422,
-      "grad_norm": 0.78520268201828,
+      "grad_norm": 1.010330080986023,
       "learning_rate": 1.3814182614927217e-05,
-      "loss": 2.0698,
+      "loss": 2.071,
       "step": 1700
     },
     {
       "epoch": 0.4467609828741623,
-      "grad_norm": 0.6703968644142151,
+      "grad_norm": 0.6752054691314697,
       "learning_rate": 1.3022788781528653e-05,
-      "loss": 2.0624,
+      "loss": 2.0636,
       "step": 1800
     },
     {
       "epoch": 0.47158103747828245,
-      "grad_norm": 0.8425729274749756,
+      "grad_norm": 0.841232180595398,
       "learning_rate": 1.220995291846777e-05,
-      "loss": 2.0517,
+      "loss": 2.0532,
       "step": 1900
     },
     {
       "epoch": 0.4964010920824026,
-      "grad_norm": 0.7934256196022034,
+      "grad_norm": 0.7984778881072998,
       "learning_rate": 1.1381440843982634e-05,
-      "loss": 2.0424,
+      "loss": 2.0438,
       "step": 2000
     },
     {
       "epoch": 0.5212211466865228,
-      "grad_norm": 0.8032339811325073,
+      "grad_norm": 0.8068585395812988,
       "learning_rate": 1.0543129574881446e-05,
-      "loss": 2.0675,
+      "loss": 2.0687,
       "step": 2100
     },
     {
       "epoch": 0.5460412012906428,
-      "grad_norm": 0.8439006805419922,
+      "grad_norm": 0.8497598767280579,
       "learning_rate": 9.700965638162112e-06,
-      "loss": 2.0462,
+      "loss": 2.0477,
       "step": 2200
     },
     {
       "epoch": 0.570861255894763,
-      "grad_norm": 0.7389697432518005,
+      "grad_norm": 0.7474705576896667,
       "learning_rate": 8.860922889564078e-06,
-      "loss": 2.0415,
+      "loss": 2.0429,
       "step": 2300
     },
     {
       "epoch": 0.5956813104988831,
-      "grad_norm": 1.0543524026870728,
+      "grad_norm": 1.0781651735305786,
       "learning_rate": 8.028960138264857e-06,
-      "loss": 2.0374,
+      "loss": 2.0389,
       "step": 2400
     },
     {
       "epoch": 0.6205013651030032,
-      "grad_norm": 0.8730601072311401,
+      "grad_norm": 0.8750322461128235,
       "learning_rate": 7.21097887830873e-06,
-      "loss": 2.0446,
+      "loss": 2.046,
       "step": 2500
     },
     {
       "epoch": 0.6453214197071233,
-      "grad_norm": 0.9236976504325867,
+      "grad_norm": 0.9259145855903625,
       "learning_rate": 6.4127814265980095e-06,
-      "loss": 2.0229,
+      "loss": 2.0243,
       "step": 2600
     },
     {
       "epoch": 0.6701414743112435,
-      "grad_norm": 1.163818120956421,
+      "grad_norm": 1.1625196933746338,
       "learning_rate": 5.640029764393366e-06,
-      "loss": 2.0498,
+      "loss": 2.0513,
       "step": 2700
     },
     {
       "epoch": 0.6949615289153637,
-      "grad_norm": 0.8231688737869263,
+      "grad_norm": 0.8271129727363586,
       "learning_rate": 4.8982053742793025e-06,
-      "loss": 2.0217,
+      "loss": 2.0228,
       "step": 2800
     },
     {
       "epoch": 0.7197815835194837,
-      "grad_norm": 0.72389817237854,
+      "grad_norm": 0.7196031212806702,
       "learning_rate": 4.1925703574897115e-06,
-      "loss": 2.0483,
+      "loss": 2.0496,
       "step": 2900
     },
     {
       "epoch": 0.7446016381236039,
-      "grad_norm": 0.7828938364982605,
+      "grad_norm": 0.7880265712738037,
       "learning_rate": 3.528130107406099e-06,
-      "loss": 2.0134,
+      "loss": 2.0145,
       "step": 3000
     },
     {
       "epoch": 0.769421692727724,
-      "grad_norm": 0.9030676484107971,
+      "grad_norm": 0.909106433391571,
       "learning_rate": 2.909597804002603e-06,
-      "loss": 2.0427,
+      "loss": 2.0437,
       "step": 3100
     },
     {
       "epoch": 0.7942417473318442,
-      "grad_norm": 1.255286693572998,
+      "grad_norm": 1.2606161832809448,
       "learning_rate": 2.341360981094921e-06,
-      "loss": 2.043,
+      "loss": 2.0443,
       "step": 3200
     },
     {
       "epoch": 0.8190618019359642,
-      "grad_norm": 0.7835339307785034,
+      "grad_norm": 0.795652449131012,
       "learning_rate": 1.8274504035470942e-06,
-      "loss": 2.0554,
+      "loss": 2.0568,
       "step": 3300
     },
     {
       "epoch": 0.8438818565400844,
-      "grad_norm": 0.8897650241851807,
+      "grad_norm": 0.8904260993003845,
       "learning_rate": 1.3715114752043746e-06,
-      "loss": 2.0773,
+      "loss": 2.0787,
       "step": 3400
     },
     {
       "epoch": 0.8687019111442045,
-      "grad_norm": 1.0972046852111816,
+      "grad_norm": 1.0925287008285522,
       "learning_rate": 9.767783803688414e-07,
-      "loss": 2.0435,
+      "loss": 2.045,
       "step": 3500
     },
     {
       "epoch": 0.8935219657483247,
-      "grad_norm": 0.794748067855835,
+      "grad_norm": 0.799608588218689,
       "learning_rate": 6.460511422441984e-07,
-      "loss": 2.0153,
+      "loss": 2.0167,
       "step": 3600
     },
     {
       "epoch": 0.9183420203524447,
-      "grad_norm": 0.9059362411499023,
+      "grad_norm": 0.9094216227531433,
       "learning_rate": 3.8167576108468994e-07,
-      "loss": 2.0554,
+      "loss": 2.057,
       "step": 3700
     },
     {
       "epoch": 0.9431620749565649,
-      "grad_norm": 0.8400096893310547,
+      "grad_norm": 0.8395094871520996,
       "learning_rate": 1.855275729374284e-07,
-      "loss": 2.0414,
+      "loss": 2.0425,
       "step": 3800
     },
     {
       "epoch": 0.9679821295606851,
-      "grad_norm": 0.8562043309211731,
+      "grad_norm": 0.8606423735618591,
       "learning_rate": 5.89979470221802e-08,
-      "loss": 2.0197,
+      "loss": 2.0208,
       "step": 3900
     },
     {
       "epoch": 0.9928021841648051,
-      "grad_norm": 0.8947063684463501,
+      "grad_norm": 0.8908767700195312,
       "learning_rate": 2.9844161102077218e-09,
-      "loss": 2.0502,
+      "loss": 2.0512,
       "step": 4000
     },
     {
       "epoch": 1.0,
       "step": 4029,
       "total_flos": 7.32108351012864e+16,
-      "train_loss": 2.103881771554424,
-      "train_runtime": 1250.9785,
-      "train_samples_per_second": 6.441,
-      "train_steps_per_second": 3.221
+      "train_loss": 2.105005581936963,
+      "train_runtime": 1251.4031,
+      "train_samples_per_second": 6.438,
+      "train_steps_per_second": 3.22
     }
   ],
   "logging_steps": 100,