Upload 17 files

latest

@@ -1 +1 @@
-global_step80
+global_step90

trainer_state.json

@@ -1,260 +1,290 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 4.776119402985074,
+  "epoch": 4.8979591836734695,
   "eval_steps": 500,
-  "global_step": 80,
+  "global_step": 90,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.12,
+      "epoch": 0.11,
       "learning_rate": 5e-05,
-      "loss": 0.8672,
+      "loss": 0.9113,
       "step": 2
     },
     {
-      "epoch": 0.24,
+      "epoch": 0.22,
       "learning_rate": 5e-05,
-      "loss": 0.8051,
+      "loss": 0.7869,
       "step": 4
     },
     {
-      "epoch": 0.36,
+      "epoch": 0.33,
       "learning_rate": 5e-05,
-      "loss": 0.6955,
+      "loss": 0.7093,
       "step": 6
     },
     {
-      "epoch": 0.48,
+      "epoch": 0.44,
       "learning_rate": 5e-05,
-      "loss": 0.6879,
+      "loss": 0.6995,
       "step": 8
     },
     {
-      "epoch": 0.6,
+      "epoch": 0.54,
       "learning_rate": 5e-05,
-      "loss": 0.6785,
+      "loss": 0.688,
       "step": 10
     },
     {
-      "epoch": 0.72,
+      "epoch": 0.65,
       "learning_rate": 5e-05,
-      "loss": 0.6584,
+      "loss": 0.6936,
       "step": 12
     },
     {
-      "epoch": 0.84,
+      "epoch": 0.76,
       "learning_rate": 5e-05,
-      "loss": 0.646,
+      "loss": 0.6948,
       "step": 14
     },
     {
-      "epoch": 0.96,
+      "epoch": 0.87,
       "learning_rate": 5e-05,
-      "loss": 0.6585,
+      "loss": 0.7126,
       "step": 16
     },
     {
-      "epoch": 1.07,
+      "epoch": 0.98,
       "learning_rate": 5e-05,
-      "loss": 0.5556,
+      "loss": 0.6552,
       "step": 18
     },
     {
-      "epoch": 1.19,
+      "epoch": 1.09,
       "learning_rate": 5e-05,
-      "loss": 0.4459,
+      "loss": 0.5185,
       "step": 20
     },
     {
-      "epoch": 1.31,
+      "epoch": 1.2,
       "learning_rate": 5e-05,
-      "loss": 0.4237,
+      "loss": 0.4545,
       "step": 22
     },
     {
-      "epoch": 1.43,
+      "epoch": 1.31,
       "learning_rate": 5e-05,
-      "loss": 0.4039,
+      "loss": 0.4339,
       "step": 24
     },
     {
-      "epoch": 1.55,
+      "epoch": 1.41,
       "learning_rate": 5e-05,
-      "loss": 0.3943,
+      "loss": 0.4015,
       "step": 26
     },
     {
-      "epoch": 1.67,
+      "epoch": 1.52,
       "learning_rate": 5e-05,
-      "loss": 0.3858,
+      "loss": 0.3998,
       "step": 28
     },
     {
-      "epoch": 1.79,
+      "epoch": 1.63,
       "learning_rate": 5e-05,
-      "loss": 0.402,
+      "loss": 0.4085,
       "step": 30
     },
     {
-      "epoch": 1.91,
+      "epoch": 1.74,
       "learning_rate": 5e-05,
-      "loss": 0.4071,
+      "loss": 0.3968,
       "step": 32
     },
     {
-      "epoch": 2.03,
+      "epoch": 1.85,
       "learning_rate": 5e-05,
-      "loss": 0.3372,
+      "loss": 0.3707,
       "step": 34
     },
     {
-      "epoch": 2.15,
+      "epoch": 1.96,
       "learning_rate": 5e-05,
-      "loss": 0.2406,
+      "loss": 0.3904,
       "step": 36
     },
     {
-      "epoch": 2.27,
+      "epoch": 2.07,
       "learning_rate": 5e-05,
-      "loss": 0.1945,
+      "loss": 0.2891,
       "step": 38
     },
     {
-      "epoch": 2.39,
+      "epoch": 2.18,
       "learning_rate": 5e-05,
-      "loss": 0.1935,
+      "loss": 0.2091,
       "step": 40
     },
     {
-      "epoch": 2.51,
+      "epoch": 2.29,
       "learning_rate": 5e-05,
-      "loss": 0.1866,
+      "loss": 0.1942,
       "step": 42
     },
     {
-      "epoch": 2.63,
+      "epoch": 2.39,
       "learning_rate": 5e-05,
-      "loss": 0.1739,
+      "loss": 0.188,
       "step": 44
     },
     {
-      "epoch": 2.75,
+      "epoch": 2.5,
       "learning_rate": 5e-05,
-      "loss": 0.1708,
+      "loss": 0.1921,
       "step": 46
     },
     {
-      "epoch": 2.87,
+      "epoch": 2.61,
       "learning_rate": 5e-05,
-      "loss": 0.1808,
+      "loss": 0.1812,
       "step": 48
     },
     {
-      "epoch": 2.99,
+      "epoch": 2.72,
       "learning_rate": 5e-05,
-      "loss": 0.1847,
+      "loss": 0.1746,
       "step": 50
     },
     {
-      "epoch": 3.1,
+      "epoch": 2.83,
       "learning_rate": 5e-05,
-      "loss": 0.1019,
+      "loss": 0.1867,
       "step": 52
     },
     {
-      "epoch": 3.22,
+      "epoch": 2.94,
       "learning_rate": 5e-05,
-      "loss": 0.0741,
+      "loss": 0.1831,
       "step": 54
     },
     {
-      "epoch": 3.34,
+      "epoch": 3.05,
       "learning_rate": 5e-05,
-      "loss": 0.073,
+      "loss": 0.1302,
       "step": 56
     },
     {
-      "epoch": 3.46,
+      "epoch": 3.16,
       "learning_rate": 5e-05,
-      "loss": 0.0748,
+      "loss": 0.0841,
       "step": 58
     },
     {
-      "epoch": 3.58,
+      "epoch": 3.27,
       "learning_rate": 5e-05,
-      "loss": 0.07,
+      "loss": 0.0747,
       "step": 60
     },
     {
-      "epoch": 3.7,
+      "epoch": 3.37,
       "learning_rate": 5e-05,
-      "loss": 0.0692,
+      "loss": 0.0754,
       "step": 62
     },
     {
-      "epoch": 3.82,
+      "epoch": 3.48,
       "learning_rate": 5e-05,
-      "loss": 0.079,
+      "loss": 0.075,
       "step": 64
     },
     {
-      "epoch": 3.94,
+      "epoch": 3.59,
       "learning_rate": 5e-05,
-      "loss": 0.0835,
+      "loss": 0.0763,
       "step": 66
     },
     {
-      "epoch": 4.06,
+      "epoch": 3.7,
       "learning_rate": 5e-05,
-      "loss": 0.0533,
+      "loss": 0.0761,
       "step": 68
     },
     {
-      "epoch": 4.18,
+      "epoch": 3.81,
       "learning_rate": 5e-05,
-      "loss": 0.0356,
+      "loss": 0.0789,
       "step": 70
     },
     {
-      "epoch": 4.3,
+      "epoch": 3.92,
       "learning_rate": 5e-05,
-      "loss": 0.0326,
+      "loss": 0.0757,
       "step": 72
     },
     {
-      "epoch": 4.42,
+      "epoch": 4.03,
       "learning_rate": 5e-05,
-      "loss": 0.0417,
+      "loss": 0.0757,
       "step": 74
     },
     {
-      "epoch": 4.54,
+      "epoch": 4.14,
       "learning_rate": 5e-05,
-      "loss": 0.0352,
+      "loss": 0.0337,
       "step": 76
     },
     {
-      "epoch": 4.66,
+      "epoch": 4.24,
       "learning_rate": 5e-05,
-      "loss": 0.042,
+      "loss": 0.0377,
       "step": 78
     },
     {
-      "epoch": 4.78,
+      "epoch": 4.35,
       "learning_rate": 5e-05,
-      "loss": 0.0359,
+      "loss": 0.0373,
       "step": 80
+    },
+    {
+      "epoch": 4.46,
+      "learning_rate": 5e-05,
+      "loss": 0.0388,
+      "step": 82
+    },
+    {
+      "epoch": 4.57,
+      "learning_rate": 5e-05,
+      "loss": 0.0433,
+      "step": 84
+    },
+    {
+      "epoch": 4.68,
+      "learning_rate": 5e-05,
+      "loss": 0.04,
+      "step": 86
+    },
+    {
+      "epoch": 4.79,
+      "learning_rate": 5e-05,
+      "loss": 0.0487,
+      "step": 88
+    },
+    {
+      "epoch": 4.9,
+      "learning_rate": 5e-05,
+      "loss": 0.0499,
+      "step": 90
     }
   ],
   "logging_steps": 2,
-  "max_steps": 80,
+  "max_steps": 90,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 5,
-  "save_steps": 20,
-  "total_flos": 58433383956480.0,
+  "save_steps": 30,
+  "total_flos": 65847448043520.0,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:307a17178835f4dc291457cc32d36551ee702b49bbfcb565831d619d7743c969
+oid sha256:3b51f0b7428a26ea5c653b835d1af27e22f6718d9487f29295fe3e3d5922149f
 size 7224