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: 1.8816
+- Loss: 1.8814
 
 ## Model description
 
@@ -57,7 +57,7 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 1.1885        | 0.9970 | 168  | 1.8816          |
+| 1.1865        | 0.9970 | 168  | 1.8814          |
 
 
 ### Framework versions

all_results.json

@@ -1,14 +1,9 @@
 {
     "epoch": 0.9970326409495549,
-    "eval_loss": 1.881588339805603,
-    "eval_runtime": 1.5168,
-    "eval_samples": 15,
-    "eval_samples_per_second": 9.23,
-    "eval_steps_per_second": 0.659,
     "total_flos": 8.227900459906499e+17,
-    "train_loss": 1.2795698436952772,
-    "train_runtime": 2845.3749,
+    "train_loss": 1.2773614100047521,
+    "train_runtime": 661.1188,
     "train_samples": 111440,
-    "train_samples_per_second": 13.249,
-    "train_steps_per_second": 0.059
+    "train_samples_per_second": 57.022,
+    "train_steps_per_second": 0.254
 }

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 0.9970326409495549,
     "total_flos": 8.227900459906499e+17,
-    "train_loss": 1.2795698436952772,
-    "train_runtime": 2845.3749,
+    "train_loss": 1.2773614100047521,
+    "train_runtime": 661.1188,
     "train_samples": 111440,
-    "train_samples_per_second": 13.249,
-    "train_steps_per_second": 0.059
+    "train_samples_per_second": 57.022,
+    "train_steps_per_second": 0.254
 }

trainer_state.json

@@ -10,258 +10,258 @@
   "log_history": [
     {
       "epoch": 0.005934718100890208,
-      "grad_norm": 3.0403032302856445,
+      "grad_norm": 3.172546863555908,
       "learning_rate": 1.1764705882352942e-05,
-      "loss": 1.8006,
+      "loss": 1.7951,
       "step": 1
     },
     {
       "epoch": 0.02967359050445104,
-      "grad_norm": 3.026293992996216,
+      "grad_norm": 3.243614435195923,
       "learning_rate": 5.882352941176471e-05,
-      "loss": 1.8024,
+      "loss": 1.8001,
       "step": 5
     },
     {
       "epoch": 0.05934718100890208,
-      "grad_norm": 2.2480087280273438,
+      "grad_norm": 2.237574577331543,
       "learning_rate": 0.00011764705882352942,
-      "loss": 1.7257,
+      "loss": 1.7245,
       "step": 10
     },
     {
       "epoch": 0.08902077151335312,
-      "grad_norm": 2.212745428085327,
+      "grad_norm": 2.2595160007476807,
       "learning_rate": 0.00017647058823529413,
-      "loss": 1.6182,
+      "loss": 1.6176,
       "step": 15
     },
     {
       "epoch": 0.11869436201780416,
-      "grad_norm": 2.061483144760132,
+      "grad_norm": 2.2810211181640625,
       "learning_rate": 0.00019980527694749952,
-      "loss": 1.4913,
+      "loss": 1.4882,
       "step": 20
     },
     {
       "epoch": 0.14836795252225518,
-      "grad_norm": 1.8516631126403809,
+      "grad_norm": 2.004383087158203,
       "learning_rate": 0.00019861804788521493,
-      "loss": 1.3717,
+      "loss": 1.3673,
       "step": 25
     },
     {
       "epoch": 0.17804154302670624,
-      "grad_norm": 1.0224816799163818,
+      "grad_norm": 1.0031766891479492,
       "learning_rate": 0.00019636458959356316,
-      "loss": 1.3122,
+      "loss": 1.3098,
       "step": 30
     },
     {
       "epoch": 0.20771513353115728,
-      "grad_norm": 0.6997563242912292,
+      "grad_norm": 0.7017138004302979,
       "learning_rate": 0.00019306926579854821,
-      "loss": 1.2841,
+      "loss": 1.2812,
       "step": 35
     },
     {
       "epoch": 0.23738872403560832,
-      "grad_norm": 0.7542662024497986,
+      "grad_norm": 0.7703685164451599,
       "learning_rate": 0.00018876770456851877,
-      "loss": 1.2682,
+      "loss": 1.265,
       "step": 40
     },
     {
       "epoch": 0.26706231454005935,
-      "grad_norm": 0.7191608548164368,
+      "grad_norm": 0.7199541926383972,
       "learning_rate": 0.00018350641311400812,
-      "loss": 1.2583,
+      "loss": 1.2555,
       "step": 45
     },
     {
       "epoch": 0.29673590504451036,
-      "grad_norm": 0.8058047890663147,
+      "grad_norm": 0.7666878700256348,
       "learning_rate": 0.0001773422749654988,
-      "loss": 1.2459,
+      "loss": 1.2433,
       "step": 50
     },
     {
       "epoch": 0.3264094955489614,
-      "grad_norm": 0.7011504769325256,
+      "grad_norm": 1.0030404329299927,
       "learning_rate": 0.00017034193496547902,
-      "loss": 1.247,
+      "loss": 1.2443,
       "step": 55
     },
     {
       "epoch": 0.3560830860534125,
-      "grad_norm": 0.6850998997688293,
+      "grad_norm": 0.6837704181671143,
       "learning_rate": 0.00016258107872407375,
-      "loss": 1.2311,
+      "loss": 1.2282,
       "step": 60
     },
     {
       "epoch": 0.3857566765578635,
-      "grad_norm": 0.6663928627967834,
+      "grad_norm": 0.6520164608955383,
       "learning_rate": 0.00015414361432856475,
-      "loss": 1.2248,
+      "loss": 1.2225,
       "step": 65
     },
     {
       "epoch": 0.41543026706231456,
-      "grad_norm": 0.6247283816337585,
+      "grad_norm": 0.6578584909439087,
       "learning_rate": 0.00014512076515391375,
-      "loss": 1.2249,
+      "loss": 1.2227,
       "step": 70
     },
     {
       "epoch": 0.44510385756676557,
-      "grad_norm": 0.7638198733329773,
+      "grad_norm": 0.7506173849105835,
       "learning_rate": 0.00013561008358255468,
-      "loss": 1.2306,
+      "loss": 1.2281,
       "step": 75
     },
     {
       "epoch": 0.47477744807121663,
-      "grad_norm": 0.7044103145599365,
+      "grad_norm": 0.6809776425361633,
       "learning_rate": 0.0001257143962968246,
-      "loss": 1.2097,
+      "loss": 1.2078,
       "step": 80
     },
     {
       "epoch": 0.5044510385756676,
-      "grad_norm": 0.6669860482215881,
+      "grad_norm": 0.6472010016441345,
       "learning_rate": 0.00011554069254722051,
-      "loss": 1.2206,
+      "loss": 1.2185,
       "step": 85
     },
     {
       "epoch": 0.5341246290801187,
-      "grad_norm": 0.6756629347801208,
+      "grad_norm": 0.6473718881607056,
       "learning_rate": 0.00010519896741619803,
-      "loss": 1.2232,
+      "loss": 1.2216,
       "step": 90
     },
     {
       "epoch": 0.5637982195845698,
-      "grad_norm": 0.6518044471740723,
+      "grad_norm": 0.6487621068954468,
       "learning_rate": 9.480103258380198e-05,
-      "loss": 1.2115,
+      "loss": 1.2095,
       "step": 95
     },
     {
       "epoch": 0.5934718100890207,
-      "grad_norm": 0.7202388644218445,
+      "grad_norm": 0.7169174551963806,
       "learning_rate": 8.445930745277953e-05,
-      "loss": 1.2217,
+      "loss": 1.2196,
       "step": 100
     },
     {
       "epoch": 0.6231454005934718,
-      "grad_norm": 0.7034718990325928,
+      "grad_norm": 0.6917266249656677,
       "learning_rate": 7.428560370317542e-05,
-      "loss": 1.2141,
+      "loss": 1.2122,
       "step": 105
     },
     {
       "epoch": 0.6528189910979229,
-      "grad_norm": 0.7892000675201416,
+      "grad_norm": 0.6533765196800232,
       "learning_rate": 6.43899164174453e-05,
-      "loss": 1.2007,
+      "loss": 1.1987,
       "step": 110
     },
     {
       "epoch": 0.6824925816023739,
-      "grad_norm": 0.7199309468269348,
+      "grad_norm": 0.740526556968689,
       "learning_rate": 5.487923484608629e-05,
-      "loss": 1.2074,
+      "loss": 1.2059,
       "step": 115
     },
     {
       "epoch": 0.712166172106825,
-      "grad_norm": 0.7223320603370667,
+      "grad_norm": 0.7236852645874023,
       "learning_rate": 4.585638567143529e-05,
-      "loss": 1.2068,
+      "loss": 1.2049,
       "step": 120
     },
     {
       "epoch": 0.7418397626112759,
-      "grad_norm": 0.6834920644760132,
+      "grad_norm": 0.6368817090988159,
       "learning_rate": 3.741892127592625e-05,
-      "loss": 1.2158,
+      "loss": 1.2139,
       "step": 125
     },
     {
       "epoch": 0.771513353115727,
-      "grad_norm": 0.6318499445915222,
+      "grad_norm": 0.6370018124580383,
       "learning_rate": 2.9658065034520978e-05,
-      "loss": 1.2025,
+      "loss": 1.2009,
       "step": 130
     },
     {
       "epoch": 0.8011869436201781,
-      "grad_norm": 0.7108297348022461,
+      "grad_norm": 0.691111147403717,
       "learning_rate": 2.265772503450122e-05,
-      "loss": 1.1998,
+      "loss": 1.1977,
       "step": 135
     },
     {
       "epoch": 0.8308605341246291,
-      "grad_norm": 0.6781418919563293,
+      "grad_norm": 0.6910406947135925,
       "learning_rate": 1.649358688599191e-05,
-      "loss": 1.2071,
+      "loss": 1.2049,
       "step": 140
     },
     {
       "epoch": 0.8605341246290801,
-      "grad_norm": 0.6723542213439941,
+      "grad_norm": 0.669152557849884,
       "learning_rate": 1.1232295431481222e-05,
-      "loss": 1.2034,
+      "loss": 1.2018,
       "step": 145
     },
     {
       "epoch": 0.8902077151335311,
-      "grad_norm": 0.694801926612854,
+      "grad_norm": 0.6983711123466492,
       "learning_rate": 6.930734201451816e-06,
-      "loss": 1.2069,
+      "loss": 1.2049,
       "step": 150
     },
     {
       "epoch": 0.9198813056379822,
-      "grad_norm": 0.7542781233787537,
+      "grad_norm": 0.7779198288917542,
       "learning_rate": 3.6354104064368566e-06,
-      "loss": 1.2018,
+      "loss": 1.1999,
       "step": 155
     },
     {
       "epoch": 0.9495548961424333,
-      "grad_norm": 0.6740858554840088,
+      "grad_norm": 0.6764826774597168,
       "learning_rate": 1.3819521147851123e-06,
-      "loss": 1.2015,
+      "loss": 1.1995,
       "step": 160
     },
     {
       "epoch": 0.9792284866468842,
-      "grad_norm": 0.6539055109024048,
+      "grad_norm": 0.6475749611854553,
       "learning_rate": 1.947230525005006e-07,
-      "loss": 1.1885,
+      "loss": 1.1865,
       "step": 165
     },
     {
       "epoch": 0.9970326409495549,
-      "eval_loss": 1.881588339805603,
-      "eval_runtime": 1.5002,
-      "eval_samples_per_second": 9.332,
-      "eval_steps_per_second": 0.667,
+      "eval_loss": 1.8814080953598022,
+      "eval_runtime": 0.6632,
+      "eval_samples_per_second": 21.108,
+      "eval_steps_per_second": 1.508,
       "step": 168
     },
     {
       "epoch": 0.9970326409495549,
       "step": 168,
       "total_flos": 8.227900459906499e+17,
-      "train_loss": 1.2795698436952772,
-      "train_runtime": 2845.3749,
-      "train_samples_per_second": 13.249,
-      "train_steps_per_second": 0.059
+      "train_loss": 1.2773614100047521,
+      "train_runtime": 661.1188,
+      "train_samples_per_second": 57.022,
+      "train_steps_per_second": 0.254
     }
   ],
   "logging_steps": 5,