End of training

README.md

@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # persuasion_simulation_qwen2.5_7b_sft_w_promp_3epochs
 
-This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on an unknown dataset.
+This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the persuasion_simulation dataset.
 
 ## Model description
 

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 4.908108108108108,
-    "total_flos": 8436467998720.0,
-    "train_loss": 0.7619479303774627,
-    "train_runtime": 1795.6366,
-    "train_samples_per_second": 2.058,
-    "train_steps_per_second": 0.128
+    "epoch": 2.9513513513513514,
+    "total_flos": 5071563390976.0,
+    "train_loss": 0.982524781987287,
+    "train_runtime": 1059.4142,
+    "train_samples_per_second": 2.093,
+    "train_steps_per_second": 0.13
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 4.908108108108108,
-    "total_flos": 8436467998720.0,
-    "train_loss": 0.7619479303774627,
-    "train_runtime": 1795.6366,
-    "train_samples_per_second": 2.058,
-    "train_steps_per_second": 0.128
+    "epoch": 2.9513513513513514,
+    "total_flos": 5071563390976.0,
+    "train_loss": 0.982524781987287,
+    "train_runtime": 1059.4142,
+    "train_samples_per_second": 2.093,
+    "train_steps_per_second": 0.13
 }

trainer_state.json

@@ -2,188 +2,118 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 4.908108108108108,
+  "epoch": 2.9513513513513514,
   "eval_steps": 500,
-  "global_step": 230,
+  "global_step": 138,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.21621621621621623,
-      "grad_norm": 8.280130797797561,
-      "learning_rate": 4.347826086956522e-06,
-      "loss": 2.7155,
+      "grad_norm": 7.197075999482914,
+      "learning_rate": 7.1428571428571436e-06,
+      "loss": 2.5723,
       "step": 10
     },
     {
       "epoch": 0.43243243243243246,
-      "grad_norm": 2.4715524570417173,
-      "learning_rate": 8.695652173913044e-06,
-      "loss": 1.3465,
+      "grad_norm": 2.225334810414002,
+      "learning_rate": 9.942341621640558e-06,
+      "loss": 1.204,
       "step": 20
     },
     {
       "epoch": 0.6486486486486487,
-      "grad_norm": 2.099989136279864,
-      "learning_rate": 9.971810547786794e-06,
-      "loss": 1.0712,
+      "grad_norm": 2.124134707404298,
+      "learning_rate": 9.594789058101154e-06,
+      "loss": 1.0538,
       "step": 30
     },
     {
       "epoch": 0.8648648648648649,
-      "grad_norm": 1.8926473625262032,
-      "learning_rate": 9.834504404631032e-06,
-      "loss": 1.0268,
+      "grad_norm": 1.8972052525289154,
+      "learning_rate": 8.953878684688492e-06,
+      "loss": 1.0201,
       "step": 40
     },
     {
       "epoch": 1.0648648648648649,
-      "grad_norm": 2.064121413621776,
-      "learning_rate": 9.586056507527266e-06,
-      "loss": 0.9506,
+      "grad_norm": 2.050396090669739,
+      "learning_rate": 8.060529912738316e-06,
+      "loss": 0.9428,
       "step": 50
     },
     {
       "epoch": 1.281081081081081,
-      "grad_norm": 1.8094825922438424,
-      "learning_rate": 9.232178493644006e-06,
-      "loss": 0.8562,
+      "grad_norm": 1.711788073870433,
+      "learning_rate": 6.971779275566593e-06,
+      "loss": 0.8491,
       "step": 60
     },
     {
       "epoch": 1.4972972972972973,
-      "grad_norm": 2.050095654745026,
-      "learning_rate": 8.781005762156593e-06,
-      "loss": 0.8412,
+      "grad_norm": 2.0490846868576202,
+      "learning_rate": 5.757138887522884e-06,
+      "loss": 0.8344,
       "step": 70
     },
     {
       "epoch": 1.7135135135135136,
-      "grad_norm": 1.8352865781006955,
-      "learning_rate": 8.24291044731378e-06,
-      "loss": 0.8523,
+      "grad_norm": 1.7041228428337094,
+      "learning_rate": 4.49415839006284e-06,
+      "loss": 0.8397,
       "step": 80
     },
     {
       "epoch": 1.9297297297297298,
-      "grad_norm": 1.9667662631699199,
-      "learning_rate": 7.630262970585355e-06,
-      "loss": 0.8436,
+      "grad_norm": 1.9251426434335601,
+      "learning_rate": 3.2634737357758994e-06,
+      "loss": 0.8296,
       "step": 90
     },
     {
       "epoch": 2.1297297297297297,
-      "grad_norm": 1.7211651987438827,
-      "learning_rate": 6.957147653634198e-06,
-      "loss": 0.7678,
+      "grad_norm": 1.7433702920708603,
+      "learning_rate": 2.1436589245260375e-06,
+      "loss": 0.7802,
       "step": 100
     },
     {
       "epoch": 2.345945945945946,
-      "grad_norm": 1.7813862468072899,
-      "learning_rate": 6.2390389299645e-06,
-      "loss": 0.6588,
+      "grad_norm": 1.8519573766604274,
+      "learning_rate": 1.2062093865360458e-06,
+      "loss": 0.7006,
       "step": 110
     },
     {
       "epoch": 2.562162162162162,
-      "grad_norm": 1.874299390396019,
-      "learning_rate": 5.492445598905843e-06,
-      "loss": 0.657,
+      "grad_norm": 1.6373359351623393,
+      "learning_rate": 5.109773021462921e-07,
+      "loss": 0.6986,
       "step": 120
     },
     {
       "epoch": 2.7783783783783784,
-      "grad_norm": 1.7516461345516408,
-      "learning_rate": 4.7345313002762545e-06,
-      "loss": 0.6495,
+      "grad_norm": 1.6305332102388383,
+      "learning_rate": 1.0235029373752758e-07,
+      "loss": 0.6844,
       "step": 130
     },
     {
-      "epoch": 2.9945945945945946,
-      "grad_norm": 1.700560828577539,
-      "learning_rate": 3.982719934736832e-06,
-      "loss": 0.6521,
-      "step": 140
-    },
-    {
-      "epoch": 3.1945945945945944,
-      "grad_norm": 2.4844933595663345,
-      "learning_rate": 3.2542951009381584e-06,
-      "loss": 0.5145,
-      "step": 150
-    },
-    {
-      "epoch": 3.410810810810811,
-      "grad_norm": 1.788000241563165,
-      "learning_rate": 2.566002758108256e-06,
-      "loss": 0.4596,
-      "step": 160
-    },
-    {
-      "epoch": 3.627027027027027,
-      "grad_norm": 1.7733098076103213,
-      "learning_rate": 1.933666248581418e-06,
-      "loss": 0.447,
-      "step": 170
-    },
-    {
-      "epoch": 3.8432432432432435,
-      "grad_norm": 1.88196318426097,
-      "learning_rate": 1.3718225306210049e-06,
-      "loss": 0.4576,
-      "step": 180
-    },
-    {
-      "epoch": 4.043243243243243,
-      "grad_norm": 1.9793294606414797,
-      "learning_rate": 8.933879842801558e-07,
-      "loss": 0.4278,
-      "step": 190
-    },
-    {
-      "epoch": 4.2594594594594595,
-      "grad_norm": 2.043336421318033,
-      "learning_rate": 5.0936147318152e-07,
-      "loss": 0.3384,
-      "step": 200
-    },
-    {
-      "epoch": 4.475675675675676,
-      "grad_norm": 1.9227579071071994,
-      "learning_rate": 2.2857148861060552e-07,
-      "loss": 0.3365,
-      "step": 210
-    },
-    {
-      "epoch": 4.691891891891892,
-      "grad_norm": 1.8184136523064685,
-      "learning_rate": 5.747318889684883e-08,
-      "loss": 0.3239,
-      "step": 220
-    },
-    {
-      "epoch": 4.908108108108108,
-      "grad_norm": 1.7870900401086147,
-      "learning_rate": 0.0,
-      "loss": 0.3306,
-      "step": 230
-    },
-    {
-      "epoch": 4.908108108108108,
-      "step": 230,
-      "total_flos": 8436467998720.0,
-      "train_loss": 0.7619479303774627,
-      "train_runtime": 1795.6366,
-      "train_samples_per_second": 2.058,
-      "train_steps_per_second": 0.128
+      "epoch": 2.9513513513513514,
+      "step": 138,
+      "total_flos": 5071563390976.0,
+      "train_loss": 0.982524781987287,
+      "train_runtime": 1059.4142,
+      "train_samples_per_second": 2.093,
+      "train_steps_per_second": 0.13
     }
   ],
   "logging_steps": 10,
-  "max_steps": 230,
+  "max_steps": 138,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 3,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -197,7 +127,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 8436467998720.0,
+  "total_flos": 5071563390976.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null