End of training

all_results.json

@@ -1,11 +1,11 @@
 {
     "epoch": 3.0,
-    "eval_loss": 1.8968859910964966,
-    "eval_runtime": 68.8078,
-    "eval_samples_per_second": 234.261,
-    "eval_steps_per_second": 3.662,
-    "train_loss": 2.6023941732047637,
-    "train_runtime": 2611.7543,
-    "train_samples_per_second": 86.401,
-    "train_steps_per_second": 1.351
+    "eval_loss": 1.654771089553833,
+    "eval_runtime": 151.8615,
+    "eval_samples_per_second": 234.003,
+    "eval_steps_per_second": 3.661,
+    "train_loss": 2.2696795797151794,
+    "train_runtime": 5651.4997,
+    "train_samples_per_second": 88.029,
+    "train_steps_per_second": 1.376
 }

eval_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 3.0,
-    "eval_loss": 1.8968859910964966,
-    "eval_runtime": 68.8078,
-    "eval_samples_per_second": 234.261,
-    "eval_steps_per_second": 3.662
+    "eval_loss": 1.654771089553833,
+    "eval_runtime": 151.8615,
+    "eval_samples_per_second": 234.003,
+    "eval_steps_per_second": 3.661
 }

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 3.0,
-    "train_loss": 2.6023941732047637,
-    "train_runtime": 2611.7543,
-    "train_samples_per_second": 86.401,
-    "train_steps_per_second": 1.351
+    "train_loss": 2.2696795797151794,
+    "train_runtime": 5651.4997,
+    "train_samples_per_second": 88.029,
+    "train_steps_per_second": 1.376
 }

trainer_state.json

@@ -2,66 +2,114 @@
   "best_metric": null,
   "best_model_checkpoint": null,
   "epoch": 3.0,
-  "global_step": 3528,
+  "global_step": 7776,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.43,
-      "learning_rate": 4.291383219954649e-05,
-      "loss": 3.7621,
+      "epoch": 0.19,
+      "learning_rate": 4.6784979423868314e-05,
+      "loss": 3.7829,
       "step": 500
     },
     {
-      "epoch": 0.85,
-      "learning_rate": 3.5827664399092974e-05,
-      "loss": 3.2286,
+      "epoch": 0.39,
+      "learning_rate": 4.3569958847736625e-05,
+      "loss": 3.3678,
       "step": 1000
     },
     {
-      "epoch": 1.28,
-      "learning_rate": 2.8741496598639456e-05,
-      "loss": 2.7611,
+      "epoch": 0.58,
+      "learning_rate": 4.035493827160494e-05,
+      "loss": 2.9585,
       "step": 1500
     },
     {
-      "epoch": 1.7,
-      "learning_rate": 2.1655328798185942e-05,
-      "loss": 2.497,
+      "epoch": 0.77,
+      "learning_rate": 3.7139917695473254e-05,
+      "loss": 2.6975,
       "step": 2000
     },
     {
-      "epoch": 2.13,
-      "learning_rate": 1.4569160997732428e-05,
-      "loss": 2.2218,
+      "epoch": 0.96,
+      "learning_rate": 3.3924897119341565e-05,
+      "loss": 2.5295,
       "step": 2500
     },
     {
-      "epoch": 2.55,
-      "learning_rate": 7.482993197278912e-06,
-      "loss": 1.9671,
+      "epoch": 1.16,
+      "learning_rate": 3.0709876543209876e-05,
+      "loss": 2.3339,
       "step": 3000
     },
     {
-      "epoch": 2.98,
-      "learning_rate": 3.9682539682539683e-07,
-      "loss": 1.8285,
+      "epoch": 1.35,
+      "learning_rate": 2.7494855967078194e-05,
+      "loss": 2.2302,
       "step": 3500
     },
+    {
+      "epoch": 1.54,
+      "learning_rate": 2.4279835390946505e-05,
+      "loss": 2.1406,
+      "step": 4000
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 2.1064814814814816e-05,
+      "loss": 2.0496,
+      "step": 4500
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.784979423868313e-05,
+      "loss": 1.979,
+      "step": 5000
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 1.463477366255144e-05,
+      "loss": 1.8329,
+      "step": 5500
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 1.1419753086419753e-05,
+      "loss": 1.7135,
+      "step": 6000
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 8.204732510288066e-06,
+      "loss": 1.6597,
+      "step": 6500
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 4.989711934156379e-06,
+      "loss": 1.615,
+      "step": 7000
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 1.7746913580246913e-06,
+      "loss": 1.5626,
+      "step": 7500
+    },
     {
       "epoch": 3.0,
-      "step": 3528,
-      "total_flos": 3.012910885296e+16,
-      "train_loss": 2.6023941732047637,
-      "train_runtime": 2611.7543,
-      "train_samples_per_second": 86.401,
-      "train_steps_per_second": 1.351
+      "step": 7776,
+      "total_flos": 6.642431273088e+16,
+      "train_loss": 2.2696795797151794,
+      "train_runtime": 5651.4997,
+      "train_samples_per_second": 88.029,
+      "train_steps_per_second": 1.376
     }
   ],
-  "max_steps": 3528,
+  "max_steps": 7776,
   "num_train_epochs": 3,
-  "total_flos": 3.012910885296e+16,
+  "total_flos": 6.642431273088e+16,
   "trial_name": null,
   "trial_params": null
 }