End of training

all_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 100.0,
-    "eval_loss": 5.574370384216309,
-    "eval_runtime": 31.2053,
-    "eval_samples_per_second": 311.326,
-    "eval_steps_per_second": 7.082,
-    "total_flos": 2.3359629906138624e+18,
-    "train_loss": 0.0,
-    "train_runtime": 3.5615,
-    "train_samples_per_second": 1660250.627,
-    "train_steps_per_second": 41528.026
+    "epoch": 150.0,
+    "eval_loss": 5.543883323669434,
+    "eval_runtime": 30.8465,
+    "eval_samples_per_second": 314.947,
+    "eval_steps_per_second": 7.165,
+    "total_flos": 3.5039444859207936e+18,
+    "train_loss": 0.007636452434781384,
+    "train_runtime": 24973.9811,
+    "train_samples_per_second": 355.144,
+    "train_steps_per_second": 8.883
 }

eval_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 100.0,
-    "eval_loss": 5.574370384216309,
-    "eval_runtime": 31.2053,
-    "eval_samples_per_second": 311.326,
-    "eval_steps_per_second": 7.082
+    "epoch": 150.0,
+    "eval_loss": 5.543883323669434,
+    "eval_runtime": 30.8465,
+    "eval_samples_per_second": 314.947,
+    "eval_steps_per_second": 7.165
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 100.0,
-    "total_flos": 2.3359629906138624e+18,
-    "train_loss": 0.0,
-    "train_runtime": 3.5615,
-    "train_samples_per_second": 1660250.627,
-    "train_steps_per_second": 41528.026
+    "epoch": 150.0,
+    "total_flos": 3.5039444859207936e+18,
+    "train_loss": 0.007636452434781384,
+    "train_runtime": 24973.9811,
+    "train_samples_per_second": 355.144,
+    "train_steps_per_second": 8.883
 }

trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 100.0,
+  "epoch": 150.0,
   "eval_steps": 14790,
-  "global_step": 147900,
+  "global_step": 221850,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -159,19 +159,94 @@
       "step": 147900
     },
     {
-      "epoch": 100.0,
-      "step": 147900,
-      "total_flos": 2.3359629906138624e+18,
-      "train_loss": 0.0,
-      "train_runtime": 3.5615,
-      "train_samples_per_second": 1660250.627,
-      "train_steps_per_second": 41528.026
+      "epoch": 110.0,
+      "grad_norm": 0.292227566242218,
+      "learning_rate": 2.6701825557809337e-06,
+      "loss": 0.0243,
+      "step": 162690
+    },
+    {
+      "epoch": 110.0,
+      "eval_loss": 5.654893398284912,
+      "eval_runtime": 30.718,
+      "eval_samples_per_second": 316.265,
+      "eval_steps_per_second": 7.194,
+      "step": 162690
+    },
+    {
+      "epoch": 120.0,
+      "grad_norm": 0.000675542454700917,
+      "learning_rate": 2.0038314176245213e-06,
+      "loss": 0.024,
+      "step": 177480
+    },
+    {
+      "epoch": 120.0,
+      "eval_loss": 5.59971284866333,
+      "eval_runtime": 30.8062,
+      "eval_samples_per_second": 315.359,
+      "eval_steps_per_second": 7.174,
+      "step": 177480
+    },
+    {
+      "epoch": 130.0,
+      "grad_norm": 0.00018167876987718046,
+      "learning_rate": 1.3374802794681092e-06,
+      "loss": 0.0227,
+      "step": 192270
+    },
+    {
+      "epoch": 130.0,
+      "eval_loss": 5.560417175292969,
+      "eval_runtime": 30.9576,
+      "eval_samples_per_second": 313.816,
+      "eval_steps_per_second": 7.139,
+      "step": 192270
+    },
+    {
+      "epoch": 140.0,
+      "grad_norm": 0.0007145697018131614,
+      "learning_rate": 6.711291413116972e-07,
+      "loss": 0.0219,
+      "step": 207060
+    },
+    {
+      "epoch": 140.0,
+      "eval_loss": 5.54286527633667,
+      "eval_runtime": 31.0413,
+      "eval_samples_per_second": 312.97,
+      "eval_steps_per_second": 7.12,
+      "step": 207060
+    },
+    {
+      "epoch": 150.0,
+      "grad_norm": 0.0009066470083780587,
+      "learning_rate": 4.778003155285103e-09,
+      "loss": 0.0217,
+      "step": 221850
+    },
+    {
+      "epoch": 150.0,
+      "eval_loss": 5.543883323669434,
+      "eval_runtime": 30.764,
+      "eval_samples_per_second": 315.791,
+      "eval_steps_per_second": 7.184,
+      "step": 221850
+    },
+    {
+      "epoch": 150.0,
+      "step": 221850,
+      "total_flos": 3.5039444859207936e+18,
+      "train_loss": 0.007636452434781384,
+      "train_runtime": 24973.9811,
+      "train_samples_per_second": 355.144,
+      "train_steps_per_second": 8.883
     }
   ],
   "logging_steps": 14790,
-  "max_steps": 147900,
+  "max_steps": 221850,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 100,
+  "num_train_epochs": 150,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -185,7 +260,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.3359629906138624e+18,
+  "total_flos": 3.5039444859207936e+18,
   "train_batch_size": 40,
   "trial_name": null,
   "trial_params": null