End of training

README.md

@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # persuasion_simulation_llama3.1_8b_lora_w_promp_10epochs
 
-This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on the persuasion_simulation dataset.
 
 ## Model description
 

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 4.908108108108108,
-    "total_flos": 33155136552960.0,
-    "train_loss": 2.138509211332902,
-    "train_runtime": 1850.4628,
-    "train_samples_per_second": 1.997,
+    "epoch": 9.8,
+    "total_flos": 66249434202112.0,
+    "train_loss": 2.112510855301567,
+    "train_runtime": 3704.1785,
+    "train_samples_per_second": 1.995,
     "train_steps_per_second": 0.124
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 4.908108108108108,
-    "total_flos": 33155136552960.0,
-    "train_loss": 2.138509211332902,
-    "train_runtime": 1850.4628,
-    "train_samples_per_second": 1.997,
+    "epoch": 9.8,
+    "total_flos": 66249434202112.0,
+    "train_loss": 2.112510855301567,
+    "train_runtime": 3704.1785,
+    "train_samples_per_second": 1.995,
     "train_steps_per_second": 0.124
 }

trainer_state.json

@@ -2,188 +2,349 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 4.908108108108108,
+  "epoch": 9.8,
   "eval_steps": 500,
-  "global_step": 230,
+  "global_step": 460,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.21621621621621623,
-      "grad_norm": 5.058993962905337,
-      "learning_rate": 2.173913043478261e-05,
-      "loss": 3.2922,
+      "grad_norm": 5.13529920590367,
+      "learning_rate": 4.347826086956522e-06,
+      "loss": 3.3574,
       "step": 10
     },
     {
       "epoch": 0.43243243243243246,
-      "grad_norm": 0.82106274957726,
-      "learning_rate": 4.347826086956522e-05,
-      "loss": 2.4785,
+      "grad_norm": 4.393430614338513,
+      "learning_rate": 8.695652173913044e-06,
+      "loss": 3.1883,
       "step": 20
     },
     {
       "epoch": 0.6486486486486487,
-      "grad_norm": 0.5924213510403318,
-      "learning_rate": 4.9859052738933966e-05,
-      "loss": 2.3045,
+      "grad_norm": 1.936708315729194,
+      "learning_rate": 1.3043478260869566e-05,
+      "loss": 2.7357,
       "step": 30
     },
     {
       "epoch": 0.8648648648648649,
-      "grad_norm": 0.4732982563249772,
-      "learning_rate": 4.9172522023155154e-05,
-      "loss": 2.2538,
+      "grad_norm": 0.8024909758301331,
+      "learning_rate": 1.739130434782609e-05,
+      "loss": 2.3914,
       "step": 40
     },
     {
       "epoch": 1.0648648648648649,
-      "grad_norm": 0.6041995035570952,
-      "learning_rate": 4.793028253763633e-05,
-      "loss": 2.2724,
+      "grad_norm": 0.6678049807544166,
+      "learning_rate": 1.9995393663024054e-05,
+      "loss": 2.3649,
       "step": 50
     },
     {
       "epoch": 1.281081081081081,
-      "grad_norm": 0.4995551008010827,
-      "learning_rate": 4.616089246822003e-05,
-      "loss": 2.1478,
+      "grad_norm": 0.5228602571380523,
+      "learning_rate": 1.9943621095573588e-05,
+      "loss": 2.2259,
       "step": 60
     },
     {
       "epoch": 1.4972972972972973,
-      "grad_norm": 0.5779223281194174,
-      "learning_rate": 4.390502881078296e-05,
-      "loss": 2.1726,
+      "grad_norm": 0.5561612233267006,
+      "learning_rate": 1.9834617016337424e-05,
+      "loss": 2.2491,
       "step": 70
     },
     {
       "epoch": 1.7135135135135136,
-      "grad_norm": 0.5455419915694549,
-      "learning_rate": 4.12145522365689e-05,
-      "loss": 2.1121,
+      "grad_norm": 0.5023960303554494,
+      "learning_rate": 1.9669008809262064e-05,
+      "loss": 2.1711,
       "step": 80
     },
     {
       "epoch": 1.9297297297297298,
-      "grad_norm": 0.6049110748406865,
-      "learning_rate": 3.815131485292678e-05,
-      "loss": 2.1814,
+      "grad_norm": 0.5205652400510364,
+      "learning_rate": 1.944774964904754e-05,
+      "loss": 2.2428,
       "step": 90
     },
     {
       "epoch": 2.1297297297297297,
-      "grad_norm": 0.5584155705576188,
-      "learning_rate": 3.478573826817099e-05,
-      "loss": 2.0904,
+      "grad_norm": 0.5104196136833851,
+      "learning_rate": 1.917211301505453e-05,
+      "loss": 2.1657,
       "step": 100
     },
     {
       "epoch": 2.345945945945946,
-      "grad_norm": 0.6598688547759153,
-      "learning_rate": 3.11951946498225e-05,
-      "loss": 2.107,
+      "grad_norm": 0.6550508381987725,
+      "learning_rate": 1.8843685361665724e-05,
+      "loss": 2.201,
       "step": 110
     },
     {
       "epoch": 2.562162162162162,
-      "grad_norm": 0.7981397754825942,
-      "learning_rate": 2.7462227994529217e-05,
-      "loss": 2.0745,
+      "grad_norm": 0.6813015740148621,
+      "learning_rate": 1.8464356987288012e-05,
+      "loss": 2.1659,
       "step": 120
     },
     {
       "epoch": 2.7783783783783784,
-      "grad_norm": 0.863490584940023,
-      "learning_rate": 2.3672656501381272e-05,
-      "loss": 2.0476,
+      "grad_norm": 0.8109529839513377,
+      "learning_rate": 1.8036311154549783e-05,
+      "loss": 2.1377,
       "step": 130
     },
     {
       "epoch": 2.9945945945945946,
-      "grad_norm": 0.7964857580667576,
-      "learning_rate": 1.991359967368416e-05,
-      "loss": 2.0155,
+      "grad_norm": 0.7034478566479526,
+      "learning_rate": 1.7562011524313187e-05,
+      "loss": 2.1044,
       "step": 140
     },
     {
       "epoch": 3.1945945945945944,
-      "grad_norm": 0.8711798298759336,
-      "learning_rate": 1.6271475504690792e-05,
-      "loss": 2.0178,
+      "grad_norm": 0.6899492216672011,
+      "learning_rate": 1.7044187975826126e-05,
+      "loss": 2.1374,
       "step": 150
     },
     {
       "epoch": 3.410810810810811,
-      "grad_norm": 0.9289052122197778,
-      "learning_rate": 1.2830013790541279e-05,
-      "loss": 1.9744,
+      "grad_norm": 0.7567233449152049,
+      "learning_rate": 1.648582089462756e-05,
+      "loss": 2.0918,
       "step": 160
     },
     {
       "epoch": 3.627027027027027,
-      "grad_norm": 0.956797944384188,
-      "learning_rate": 9.668331242907089e-06,
-      "loss": 1.9704,
+      "grad_norm": 0.7796712512408794,
+      "learning_rate": 1.589012401863864e-05,
+      "loss": 2.0918,
       "step": 170
     },
     {
       "epoch": 3.8432432432432435,
-      "grad_norm": 0.9701630782260972,
-      "learning_rate": 6.859112653105024e-06,
-      "loss": 1.9838,
+      "grad_norm": 0.8165289085575146,
+      "learning_rate": 1.526052594117071e-05,
+      "loss": 2.1006,
       "step": 180
     },
     {
       "epoch": 4.043243243243243,
-      "grad_norm": 1.0442276197366696,
-      "learning_rate": 4.4669399214007785e-06,
-      "loss": 1.9445,
+      "grad_norm": 0.8774581999436206,
+      "learning_rate": 1.4600650377311523e-05,
+      "loss": 2.0631,
       "step": 190
     },
     {
       "epoch": 4.2594594594594595,
-      "grad_norm": 0.9943834280541437,
-      "learning_rate": 2.5468073659076e-06,
-      "loss": 1.8969,
+      "grad_norm": 0.8830217676503004,
+      "learning_rate": 1.3914295307268396e-05,
+      "loss": 2.0103,
       "step": 200
     },
     {
       "epoch": 4.475675675675676,
-      "grad_norm": 0.9070384447511413,
-      "learning_rate": 1.1428574430530276e-06,
-      "loss": 1.966,
+      "grad_norm": 0.799477322126529,
+      "learning_rate": 1.3205411116710973e-05,
+      "loss": 2.0786,
       "step": 210
     },
     {
       "epoch": 4.691891891891892,
-      "grad_norm": 1.0267331556207562,
-      "learning_rate": 2.8736594448424415e-07,
-      "loss": 1.9238,
+      "grad_norm": 0.9950791972239992,
+      "learning_rate": 1.2478077859929e-05,
+      "loss": 2.0405,
       "step": 220
     },
     {
       "epoch": 4.908108108108108,
-      "grad_norm": 0.9813900691292579,
-      "learning_rate": 0.0,
-      "loss": 1.9579,
+      "grad_norm": 0.9716171553607408,
+      "learning_rate": 1.1736481776669307e-05,
+      "loss": 2.0715,
       "step": 230
     },
     {
-      "epoch": 4.908108108108108,
-      "step": 230,
-      "total_flos": 33155136552960.0,
-      "train_loss": 2.138509211332902,
-      "train_runtime": 1850.4628,
-      "train_samples_per_second": 1.997,
+      "epoch": 5.108108108108108,
+      "grad_norm": 1.0437517789485815,
+      "learning_rate": 1.0984891197811686e-05,
+      "loss": 2.0787,
+      "step": 240
+    },
+    {
+      "epoch": 5.324324324324325,
+      "grad_norm": 1.1284255406524875,
+      "learning_rate": 1.0227631978561057e-05,
+      "loss": 2.0428,
+      "step": 250
+    },
+    {
+      "epoch": 5.54054054054054,
+      "grad_norm": 1.2156624868382253,
+      "learning_rate": 9.469062600552509e-06,
+      "loss": 1.9815,
+      "step": 260
+    },
+    {
+      "epoch": 5.756756756756757,
+      "grad_norm": 1.075765481356295,
+      "learning_rate": 8.71354908617169e-06,
+      "loss": 1.9817,
+      "step": 270
+    },
+    {
+      "epoch": 5.972972972972973,
+      "grad_norm": 1.2019533755467047,
+      "learning_rate": 7.965439869473664e-06,
+      "loss": 1.9777,
+      "step": 280
+    },
+    {
+      "epoch": 6.172972972972973,
+      "grad_norm": 1.196361258253227,
+      "learning_rate": 7.2290407683331154e-06,
+      "loss": 2.0107,
+      "step": 290
+    },
+    {
+      "epoch": 6.389189189189189,
+      "grad_norm": 1.3849280057662594,
+      "learning_rate": 6.508590201876317e-06,
+      "loss": 1.9658,
+      "step": 300
+    },
+    {
+      "epoch": 6.605405405405405,
+      "grad_norm": 1.4118658665589596,
+      "learning_rate": 5.8082347958333625e-06,
+      "loss": 1.9686,
+      "step": 310
+    },
+    {
+      "epoch": 6.821621621621622,
+      "grad_norm": 1.3645291090723715,
+      "learning_rate": 5.132005516216512e-06,
+      "loss": 1.9171,
+      "step": 320
+    },
+    {
+      "epoch": 7.021621621621621,
+      "grad_norm": 2.530018156584274,
+      "learning_rate": 4.483794468689728e-06,
+      "loss": 1.9656,
+      "step": 330
+    },
+    {
+      "epoch": 7.237837837837838,
+      "grad_norm": 1.2812770231666408,
+      "learning_rate": 3.867332497162836e-06,
+      "loss": 1.9406,
+      "step": 340
+    },
+    {
+      "epoch": 7.454054054054054,
+      "grad_norm": 1.4355421422143646,
+      "learning_rate": 3.2861677105440335e-06,
+      "loss": 1.9101,
+      "step": 350
+    },
+    {
+      "epoch": 7.6702702702702705,
+      "grad_norm": 1.4195378042671556,
+      "learning_rate": 2.7436450612420098e-06,
+      "loss": 1.9689,
+      "step": 360
+    },
+    {
+      "epoch": 7.886486486486486,
+      "grad_norm": 1.251311305304468,
+      "learning_rate": 2.2428870929558012e-06,
+      "loss": 1.9372,
+      "step": 370
+    },
+    {
+      "epoch": 8.086486486486486,
+      "grad_norm": 1.6701725323534269,
+      "learning_rate": 1.7867759685603115e-06,
+      "loss": 1.8679,
+      "step": 380
+    },
+    {
+      "epoch": 8.302702702702703,
+      "grad_norm": 1.5053752982554525,
+      "learning_rate": 1.3779368815278648e-06,
+      "loss": 1.9392,
+      "step": 390
+    },
+    {
+      "epoch": 8.518918918918919,
+      "grad_norm": 1.354082392478539,
+      "learning_rate": 1.01872294636304e-06,
+      "loss": 1.8973,
+      "step": 400
+    },
+    {
+      "epoch": 8.735135135135135,
+      "grad_norm": 1.5413060756984076,
+      "learning_rate": 7.1120165501533e-07,
+      "loss": 1.8888,
+      "step": 410
+    },
+    {
+      "epoch": 8.951351351351352,
+      "grad_norm": 1.618741974932159,
+      "learning_rate": 4.5714297722121105e-07,
+      "loss": 1.9519,
+      "step": 420
+    },
+    {
+      "epoch": 9.151351351351352,
+      "grad_norm": 1.411723684213353,
+      "learning_rate": 2.5800917326521013e-07,
+      "loss": 1.8573,
+      "step": 430
+    },
+    {
+      "epoch": 9.367567567567567,
+      "grad_norm": 1.4018461713222472,
+      "learning_rate": 1.1494637779369766e-07,
+      "loss": 1.9243,
+      "step": 440
+    },
+    {
+      "epoch": 9.583783783783783,
+      "grad_norm": 1.504600821181125,
+      "learning_rate": 2.8778003121607834e-08,
+      "loss": 1.8928,
+      "step": 450
+    },
+    {
+      "epoch": 9.8,
+      "grad_norm": 1.3265915384266154,
+      "learning_rate": 0.0,
+      "loss": 1.922,
+      "step": 460
+    },
+    {
+      "epoch": 9.8,
+      "step": 460,
+      "total_flos": 66249434202112.0,
+      "train_loss": 2.112510855301567,
+      "train_runtime": 3704.1785,
+      "train_samples_per_second": 1.995,
       "train_steps_per_second": 0.124
     }
   ],
   "logging_steps": 10,
-  "max_steps": 230,
+  "max_steps": 460,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 10,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -197,7 +358,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 33155136552960.0,
+  "total_flos": 66249434202112.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null