Model save

README.md

@@ -27,7 +27,7 @@ print(output["generated_text"])
 
 ## Training procedure
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/wxnfifth/huggingface/runs/j55yx0w5) 
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/wxnfifth/huggingface/runs/wgko6wgr) 
 
 
 This model was trained with SFT.

all_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 0.997779422649889,
     "total_flos": 76745898196992.0,
-    "train_loss": 0.7870922902217604,
-    "train_runtime": 4743.2538,
+    "train_loss": 0.7871017519727305,
+    "train_runtime": 4713.7692,
     "train_samples": 16610,
-    "train_samples_per_second": 4.556,
+    "train_samples_per_second": 4.584,
     "train_steps_per_second": 0.071
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9c30079c59ee74a761bb370b7482619c19b48ace8a8c63e17b6b0cadc3f47833
+oid sha256:edc815fe6e6c82998edf9da2ec19888049d8c4542e361cde669b14ef26d3906e
 size 3087467144

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 0.997779422649889,
     "total_flos": 76745898196992.0,
-    "train_loss": 0.7870922902217604,
-    "train_runtime": 4743.2538,
+    "train_loss": 0.7871017519727305,
+    "train_runtime": 4713.7692,
     "train_samples": 16610,
-    "train_samples_per_second": 4.556,
+    "train_samples_per_second": 4.584,
     "train_steps_per_second": 0.071
 }

trainer_state.json

@@ -10,504 +10,504 @@
   "log_history": [
     {
       "epoch": 0.014803849000740192,
-      "grad_norm": 0.6502017974853516,
+      "grad_norm": 0.6499719023704529,
       "learning_rate": 2.9411764705882355e-06,
       "loss": 1.09,
       "step": 5
     },
     {
       "epoch": 0.029607698001480384,
-      "grad_norm": 0.3823283016681671,
+      "grad_norm": 0.38171473145484924,
       "learning_rate": 5.882352941176471e-06,
       "loss": 1.0792,
       "step": 10
     },
     {
       "epoch": 0.04441154700222058,
-      "grad_norm": 0.3939237594604492,
+      "grad_norm": 0.3942464590072632,
       "learning_rate": 8.823529411764707e-06,
       "loss": 1.0223,
       "step": 15
     },
     {
       "epoch": 0.05921539600296077,
-      "grad_norm": 0.2809950113296509,
+      "grad_norm": 0.28095921874046326,
       "learning_rate": 1.1764705882352942e-05,
       "loss": 0.9451,
       "step": 20
     },
     {
       "epoch": 0.07401924500370097,
-      "grad_norm": 0.22726929187774658,
+      "grad_norm": 0.22764872014522552,
       "learning_rate": 1.4705882352941179e-05,
       "loss": 0.9125,
       "step": 25
     },
     {
       "epoch": 0.08882309400444116,
-      "grad_norm": 0.17815199494361877,
+      "grad_norm": 0.1783059984445572,
       "learning_rate": 1.7647058823529414e-05,
       "loss": 0.893,
       "step": 30
     },
     {
       "epoch": 0.10362694300518134,
-      "grad_norm": 0.1736987829208374,
+      "grad_norm": 0.17370979487895966,
       "learning_rate": 1.9999462497359468e-05,
-      "loss": 0.8651,
+      "loss": 0.8652,
       "step": 35
     },
     {
       "epoch": 0.11843079200592153,
-      "grad_norm": 0.14923641085624695,
+      "grad_norm": 0.14947360754013062,
       "learning_rate": 1.9980655971335944e-05,
       "loss": 0.8452,
       "step": 40
     },
     {
       "epoch": 0.13323464100666174,
-      "grad_norm": 0.12440051883459091,
+      "grad_norm": 0.12460564076900482,
       "learning_rate": 1.993503206718859e-05,
       "loss": 0.8228,
       "step": 45
     },
     {
       "epoch": 0.14803849000740193,
-      "grad_norm": 0.14250266551971436,
+      "grad_norm": 0.14311614632606506,
       "learning_rate": 1.986271337340182e-05,
-      "loss": 0.8278,
+      "loss": 0.8277,
       "step": 50
     },
     {
       "epoch": 0.16284233900814213,
-      "grad_norm": 0.12134958058595657,
+      "grad_norm": 0.12113290280103683,
       "learning_rate": 1.976389420563607e-05,
       "loss": 0.8105,
       "step": 55
     },
     {
       "epoch": 0.17764618800888232,
-      "grad_norm": 0.1252334713935852,
+      "grad_norm": 0.12570306658744812,
       "learning_rate": 1.9638840084614182e-05,
-      "loss": 0.7963,
+      "loss": 0.7964,
       "step": 60
     },
     {
       "epoch": 0.19245003700962252,
-      "grad_norm": 0.12292112410068512,
+      "grad_norm": 0.12238704413175583,
       "learning_rate": 1.9487887022684336e-05,
-      "loss": 0.8063,
+      "loss": 0.8062,
       "step": 65
     },
     {
       "epoch": 0.20725388601036268,
-      "grad_norm": 0.13962922990322113,
+      "grad_norm": 0.13958358764648438,
       "learning_rate": 1.9311440620976597e-05,
-      "loss": 0.799,
+      "loss": 0.7989,
       "step": 70
     },
     {
       "epoch": 0.22205773501110287,
-      "grad_norm": 0.1244792640209198,
+      "grad_norm": 0.1243973821401596,
       "learning_rate": 1.9109974979578852e-05,
-      "loss": 0.7898,
+      "loss": 0.7899,
       "step": 75
     },
     {
       "epoch": 0.23686158401184307,
-      "grad_norm": 0.12688860297203064,
+      "grad_norm": 0.12657789885997772,
       "learning_rate": 1.8884031423660492e-05,
       "loss": 0.8185,
       "step": 80
     },
     {
       "epoch": 0.25166543301258326,
-      "grad_norm": 0.12278366833925247,
+      "grad_norm": 0.12268061190843582,
       "learning_rate": 1.8634217048966638e-05,
       "loss": 0.801,
       "step": 85
     },
     {
       "epoch": 0.2664692820133235,
-      "grad_norm": 0.11767291277647018,
+      "grad_norm": 0.1173299178481102,
       "learning_rate": 1.836120309059107e-05,
       "loss": 0.7838,
       "step": 90
     },
     {
       "epoch": 0.28127313101406365,
-      "grad_norm": 0.12599965929985046,
+      "grad_norm": 0.1265975385904312,
       "learning_rate": 1.8065723119410885e-05,
-      "loss": 0.7808,
+      "loss": 0.7809,
       "step": 95
     },
     {
       "epoch": 0.29607698001480387,
-      "grad_norm": 0.13216127455234528,
+      "grad_norm": 0.13352826237678528,
       "learning_rate": 1.77485710710289e-05,
       "loss": 0.7879,
       "step": 100
     },
     {
       "epoch": 0.29607698001480387,
-      "eval_loss": 0.80430006980896,
-      "eval_runtime": 5.8595,
-      "eval_samples_per_second": 21.845,
-      "eval_steps_per_second": 1.365,
+      "eval_loss": 0.8042058944702148,
+      "eval_runtime": 5.8168,
+      "eval_samples_per_second": 22.005,
+      "eval_steps_per_second": 1.375,
       "step": 100
     },
     {
       "epoch": 0.31088082901554404,
-      "grad_norm": 0.12097267806529999,
+      "grad_norm": 0.12055955082178116,
       "learning_rate": 1.741059911251997e-05,
       "loss": 0.7786,
       "step": 105
     },
     {
       "epoch": 0.32568467801628426,
-      "grad_norm": 0.12593290209770203,
+      "grad_norm": 0.12688305974006653,
       "learning_rate": 1.7052715352713076e-05,
-      "loss": 0.7727,
+      "loss": 0.7726,
       "step": 110
     },
     {
       "epoch": 0.3404885270170244,
-      "grad_norm": 0.12331051379442215,
+      "grad_norm": 0.12400885671377182,
       "learning_rate": 1.667588140216154e-05,
-      "loss": 0.7994,
+      "loss": 0.7996,
       "step": 115
     },
     {
       "epoch": 0.35529237601776464,
-      "grad_norm": 0.1359540820121765,
+      "grad_norm": 0.13515259325504303,
       "learning_rate": 1.628110978935756e-05,
       "loss": 0.774,
       "step": 120
     },
     {
       "epoch": 0.3700962250185048,
-      "grad_norm": 0.13382965326309204,
+      "grad_norm": 0.1352948397397995,
       "learning_rate": 1.586946124013354e-05,
       "loss": 0.7734,
       "step": 125
     },
     {
       "epoch": 0.38490007401924503,
-      "grad_norm": 0.12126770615577698,
+      "grad_norm": 0.12024685740470886,
       "learning_rate": 1.5442041827560274e-05,
-      "loss": 0.7497,
+      "loss": 0.7498,
       "step": 130
     },
     {
       "epoch": 0.3997039230199852,
-      "grad_norm": 0.1166161447763443,
+      "grad_norm": 0.11641304194927216,
       "learning_rate": 1.5000000000000002e-05,
-      "loss": 0.7607,
+      "loss": 0.7608,
       "step": 135
     },
     {
       "epoch": 0.41450777202072536,
-      "grad_norm": 0.12959471344947815,
+      "grad_norm": 0.12957507371902466,
       "learning_rate": 1.4544523495299843e-05,
       "loss": 0.7669,
       "step": 140
     },
     {
       "epoch": 0.4293116210214656,
-      "grad_norm": 0.134933203458786,
+      "grad_norm": 0.13438324630260468,
       "learning_rate": 1.4076836149416889e-05,
       "loss": 0.7829,
       "step": 145
     },
     {
       "epoch": 0.44411547002220575,
-      "grad_norm": 0.13227751851081848,
+      "grad_norm": 0.17085708677768707,
       "learning_rate": 1.3598194608050011e-05,
       "loss": 0.7677,
       "step": 150
     },
     {
       "epoch": 0.45891931902294597,
-      "grad_norm": 0.11633738875389099,
+      "grad_norm": 0.11604870110750198,
       "learning_rate": 1.3109884950114007e-05,
       "loss": 0.7567,
       "step": 155
     },
     {
       "epoch": 0.47372316802368614,
-      "grad_norm": 0.12226972728967667,
+      "grad_norm": 0.12267672270536423,
       "learning_rate": 1.2613219232128608e-05,
       "loss": 0.7568,
       "step": 160
     },
     {
       "epoch": 0.48852701702442636,
-      "grad_norm": 0.11851673573255539,
+      "grad_norm": 0.11862842738628387,
       "learning_rate": 1.2109531962807333e-05,
       "loss": 0.7583,
       "step": 165
     },
     {
       "epoch": 0.5033308660251665,
-      "grad_norm": 0.118564672768116,
+      "grad_norm": 0.1191529631614685,
       "learning_rate": 1.1600176517318742e-05,
       "loss": 0.7631,
       "step": 170
     },
     {
       "epoch": 0.5181347150259067,
-      "grad_norm": 0.12198054790496826,
+      "grad_norm": 0.12247402966022491,
       "learning_rate": 1.1086521500854746e-05,
       "loss": 0.75,
       "step": 175
     },
     {
       "epoch": 0.532938564026647,
-      "grad_norm": 0.1204327940940857,
+      "grad_norm": 0.12190617620944977,
       "learning_rate": 1.0569947071276847e-05,
       "loss": 0.7708,
       "step": 180
     },
     {
       "epoch": 0.5477424130273871,
-      "grad_norm": 0.13036802411079407,
+      "grad_norm": 0.13005486130714417,
       "learning_rate": 1.0051841230721065e-05,
       "loss": 0.764,
       "step": 185
     },
     {
       "epoch": 0.5625462620281273,
-      "grad_norm": 0.13102592527866364,
+      "grad_norm": 0.13146714866161346,
       "learning_rate": 9.533596096125826e-06,
-      "loss": 0.7705,
+      "loss": 0.7706,
       "step": 190
     },
     {
       "epoch": 0.5773501110288675,
-      "grad_norm": 0.12162639945745468,
+      "grad_norm": 0.1219043880701065,
       "learning_rate": 9.016604158703654e-06,
       "loss": 0.7444,
       "step": 195
     },
     {
       "epoch": 0.5921539600296077,
-      "grad_norm": 0.13102519512176514,
+      "grad_norm": 0.13030683994293213,
       "learning_rate": 8.502254542407186e-06,
       "loss": 0.7423,
       "step": 200
     },
     {
       "epoch": 0.5921539600296077,
-      "eval_loss": 0.7782641053199768,
-      "eval_runtime": 5.9255,
-      "eval_samples_per_second": 21.601,
-      "eval_steps_per_second": 1.35,
+      "eval_loss": 0.7782207131385803,
+      "eval_runtime": 5.771,
+      "eval_samples_per_second": 22.18,
+      "eval_steps_per_second": 1.386,
       "step": 200
     },
     {
       "epoch": 0.6069578090303479,
-      "grad_norm": 0.11822319775819778,
+      "grad_norm": 0.1178601086139679,
       "learning_rate": 7.991929271442817e-06,
       "loss": 0.7461,
       "step": 205
     },
     {
       "epoch": 0.6217616580310881,
-      "grad_norm": 0.11643572896718979,
+      "grad_norm": 0.1156802773475647,
       "learning_rate": 7.48699955686089e-06,
       "loss": 0.7483,
       "step": 210
     },
     {
       "epoch": 0.6365655070318282,
-      "grad_norm": 0.11460445076227188,
+      "grad_norm": 0.11491943150758743,
       "learning_rate": 6.988822112200157e-06,
-      "loss": 0.7567,
+      "loss": 0.7566,
       "step": 215
     },
     {
       "epoch": 0.6513693560325685,
-      "grad_norm": 0.12710332870483398,
+      "grad_norm": 0.12633360922336578,
       "learning_rate": 6.498735508086094e-06,
       "loss": 0.7597,
       "step": 220
     },
     {
       "epoch": 0.6661732050333087,
-      "grad_norm": 0.11362723261117935,
+      "grad_norm": 0.11288498342037201,
       "learning_rate": 6.018056575578075e-06,
-      "loss": 0.7537,
+      "loss": 0.7536,
       "step": 225
     },
     {
       "epoch": 0.6809770540340488,
-      "grad_norm": 0.10726357251405716,
+      "grad_norm": 0.10684435814619064,
       "learning_rate": 5.548076867929331e-06,
       "loss": 0.7503,
       "step": 230
     },
     {
       "epoch": 0.695780903034789,
-      "grad_norm": 0.11590282618999481,
+      "grad_norm": 0.11590610444545746,
       "learning_rate": 5.090059190266779e-06,
       "loss": 0.7384,
       "step": 235
     },
     {
       "epoch": 0.7105847520355293,
-      "grad_norm": 0.10790540277957916,
+      "grad_norm": 0.10783125460147858,
       "learning_rate": 4.645234206515171e-06,
-      "loss": 0.7436,
+      "loss": 0.7435,
       "step": 240
     },
     {
       "epoch": 0.7253886010362695,
-      "grad_norm": 0.11345735192298889,
+      "grad_norm": 0.11286304891109467,
       "learning_rate": 4.214797132682597e-06,
       "loss": 0.7401,
       "step": 245
     },
     {
       "epoch": 0.7401924500370096,
-      "grad_norm": 0.11736641824245453,
+      "grad_norm": 0.11659123748540878,
       "learning_rate": 3.799904525392251e-06,
       "loss": 0.747,
       "step": 250
     },
     {
       "epoch": 0.7549962990377498,
-      "grad_norm": 0.11365947127342224,
+      "grad_norm": 0.11270508170127869,
       "learning_rate": 3.401671174289469e-06,
       "loss": 0.7371,
       "step": 255
     },
     {
       "epoch": 0.7698001480384901,
-      "grad_norm": 0.11066755652427673,
+      "grad_norm": 0.11058636754751205,
       "learning_rate": 3.021167106673928e-06,
-      "loss": 0.7531,
+      "loss": 0.7532,
       "step": 260
     },
     {
       "epoch": 0.7846039970392302,
-      "grad_norm": 0.10806834697723389,
+      "grad_norm": 0.10753703117370605,
       "learning_rate": 2.6594147124053983e-06,
-      "loss": 0.742,
+      "loss": 0.7419,
       "step": 265
     },
     {
       "epoch": 0.7994078460399704,
-      "grad_norm": 0.1102728471159935,
+      "grad_norm": 0.10991961508989334,
       "learning_rate": 2.317385996808195e-06,
       "loss": 0.7536,
       "step": 270
     },
     {
       "epoch": 0.8142116950407106,
-      "grad_norm": 0.10305000841617584,
+      "grad_norm": 0.10330849140882492,
       "learning_rate": 1.9959999689556407e-06,
       "loss": 0.7463,
       "step": 275
     },
     {
       "epoch": 0.8290155440414507,
-      "grad_norm": 0.10573720186948776,
+      "grad_norm": 0.10575641691684723,
       "learning_rate": 1.6961201723520248e-06,
       "loss": 0.732,
       "step": 280
     },
     {
       "epoch": 0.843819393042191,
-      "grad_norm": 0.10039414465427399,
+      "grad_norm": 0.10046100616455078,
       "learning_rate": 1.4185523646469822e-06,
       "loss": 0.757,
       "step": 285
     },
     {
       "epoch": 0.8586232420429312,
-      "grad_norm": 0.10779191553592682,
+      "grad_norm": 0.10767965763807297,
       "learning_rate": 1.1640423526166987e-06,
       "loss": 0.7348,
       "step": 290
     },
     {
       "epoch": 0.8734270910436713,
-      "grad_norm": 0.1016748696565628,
+      "grad_norm": 0.10156064480543137,
       "learning_rate": 9.332739882292752e-07,
       "loss": 0.7608,
       "step": 295
     },
     {
       "epoch": 0.8882309400444115,
-      "grad_norm": 0.10107531398534775,
+      "grad_norm": 0.10109930485486984,
       "learning_rate": 7.268673311786378e-07,
-      "loss": 0.7507,
+      "loss": 0.7508,
       "step": 300
     },
     {
       "epoch": 0.8882309400444115,
-      "eval_loss": 0.7698361873626709,
-      "eval_runtime": 5.8048,
-      "eval_samples_per_second": 22.051,
-      "eval_steps_per_second": 1.378,
+      "eval_loss": 0.7698501348495483,
+      "eval_runtime": 5.8951,
+      "eval_samples_per_second": 21.713,
+      "eval_steps_per_second": 1.357,
       "step": 300
     },
     {
       "epoch": 0.9030347890451518,
-      "grad_norm": 0.09836234152317047,
+      "grad_norm": 0.0981217697262764,
       "learning_rate": 5.453769828241872e-07,
       "loss": 0.7343,
       "step": 305
     },
     {
       "epoch": 0.9178386380458919,
-      "grad_norm": 0.10397884249687195,
+      "grad_norm": 0.10373206436634064,
       "learning_rate": 3.8929059601275463e-07,
       "loss": 0.7668,
       "step": 310
     },
     {
       "epoch": 0.9326424870466321,
-      "grad_norm": 0.09893100708723068,
+      "grad_norm": 0.09866725653409958,
       "learning_rate": 2.5902756478688674e-07,
       "loss": 0.749,
       "step": 315
     },
     {
       "epoch": 0.9474463360473723,
-      "grad_norm": 0.1005384624004364,
+      "grad_norm": 0.10060535371303558,
       "learning_rate": 1.5493789750014032e-07,
       "loss": 0.7509,
       "step": 320
     },
     {
       "epoch": 0.9622501850481125,
-      "grad_norm": 0.10756874084472656,
+      "grad_norm": 0.10740893334150314,
       "learning_rate": 7.730127636723539e-08,
       "loss": 0.7315,
       "step": 325
     },
     {
       "epoch": 0.9770540340488527,
-      "grad_norm": 0.1049792617559433,
+      "grad_norm": 0.10483107715845108,
       "learning_rate": 2.6326305976001054e-08,
       "loss": 0.7362,
       "step": 330
     },
     {
       "epoch": 0.9918578830495929,
-      "grad_norm": 0.11218578368425369,
+      "grad_norm": 0.11270838230848312,
       "learning_rate": 2.149952780321485e-09,
-      "loss": 0.7576,
+      "loss": 0.7575,
       "step": 335
     },
     {
       "epoch": 0.997779422649889,
       "step": 337,
       "total_flos": 76745898196992.0,
-      "train_loss": 0.7870922902217604,
-      "train_runtime": 4743.2538,
-      "train_samples_per_second": 4.556,
+      "train_loss": 0.7871017519727305,
+      "train_runtime": 4713.7692,
+      "train_samples_per_second": 4.584,
       "train_steps_per_second": 0.071
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0516457824250b3c72e8f3cba31e7ed9ce0733070d5423a99e9bb91778da8840
+oid sha256:7e03557d5ccbfcde45badb3bdd2307a19e7b1d15f3e6ea0f661a780e2ee06c2c
 size 7352